Search results for: data driven decision making

Authors: Pei Yi Lin

Abstract:

Objective: The objective of this study is to use machine learning methods to build an early prediction classifier model for acute delirium to improve the quality of medical care for intensive care patients. Background: Delirium is a common acute and sudden disturbance of consciousness in critically ill patients. After the occurrence, it is easy to prolong the length of hospital stay and increase medical costs and mortality. In 2021, the incidence of delirium in the intensive care unit of internal medicine was as high as 59.78%, which indirectly prolonged the average length of hospital stay by 8.28 days, and the mortality rate is about 2.22% in the past three years. Therefore, it is expected to build a delirium prediction classifier through big data analysis and machine learning methods to detect delirium early. Method: This study is a retrospective study, using the artificial intelligence big data database to extract the characteristic factors related to delirium in intensive care unit patients and let the machine learn. The study included patients aged over 20 years old who were admitted to the intensive care unit between May 1, 2022, and December 31, 2022, excluding GCS assessment <4 points, admission to ICU for less than 24 hours, and CAM-ICU evaluation. The CAMICU delirium assessment results every 8 hours within 30 days of hospitalization are regarded as an event, and the cumulative data from ICU admission to the prediction time point are extracted to predict the possibility of delirium occurring in the next 8 hours, and collect a total of 63,754 research case data, extract 12 feature selections to train the model, including age, sex, average ICU stay hours, visual and auditory abnormalities, RASS assessment score, APACHE-II Score score, number of invasive catheters indwelling, restraint and sedative and hypnotic drugs. Through feature data cleaning, processing and KNN interpolation method supplementation, a total of 54595 research case events were extracted to provide machine learning model analysis, using the research events from May 01 to November 30, 2022, as the model training data, 80% of which is the training set for model training, and 20% for the internal verification of the verification set, and then from December 01 to December 2022 The CU research event on the 31st is an external verification set data, and finally the model inference and performance evaluation are performed, and then the model has trained again by adjusting the model parameters. Results: In this study, XG Boost, Random Forest, Logistic Regression, and Decision Tree were used to analyze and compare four machine learning models. The average accuracy rate of internal verification was highest in Random Forest (AUC=0.86), and the average accuracy rate of external verification was in Random Forest and XG Boost was the highest, AUC was 0.86, and the average accuracy of cross-validation was the highest in Random Forest (ACC=0.77). Conclusion: Clinically, medical staff usually conduct CAM-ICU assessments at the bedside of critically ill patients in clinical practice, but there is a lack of machine learning classification methods to assist ICU patients in real-time assessment, resulting in the inability to provide more objective and continuous monitoring data to assist Clinical staff can more accurately identify and predict the occurrence of delirium in patients. It is hoped that the development and construction of predictive models through machine learning can predict delirium early and immediately, make clinical decisions at the best time, and cooperate with PADIS delirium care measures to provide individualized non-drug interventional care measures to maintain patient safety, and then Improve the quality of care.

Keywords: critically ill patients, machine learning methods, delirium prediction, classifier model

Procedia PDF Downloads 80

Authors: Sufrin Hannan, Budi Suharjo, Rita Nurmalina, Kirbrandoko

Abstract:

The challenge for independent surveyor service companies now is growing with increasing uncertainty in business. Protection from the government for domestic independent surveyor industry from competitor attack, such as entering the global surveyors to Indonesia also no longer exists. Therefore, building customer loyalty becomes very important to create a long-term relationship between an independent surveyor with its customers. This study aims to develop a model that can be used to build customer loyalty by looking at various factors that determine customer loyalty, especially on independent surveyors for coal inspection in Indonesia. The development of this model uses the relationship marketing approach. Testing of the hypothesis is done by testing the variables that determine customer loyalty, either directly or indirectly, which amounted to 10 variables. The data were collected from 200 questionnaires filled by independent surveyor company decision makers from 51 exporting companies and coal trading companies in Indonesia and analyzed using Structural Equation Model (SEM). The results show that customer loyalty of independent surveyors is influenced by customer satisfaction, trust, switching-barrier, and relationship-bond. Research on customer satisfaction shows that customer satisfaction is influenced by the perceived quality and perceived value, while perceived quality is influenced by reliability, assurance, responsiveness, and empathy.

Keywords: relationship marketing, customer loyalty, customer satisfaction, switching barriers, relationship bonds

Procedia PDF Downloads 172

Authors: Yina F. Muñoz, Alexander Paz, Hanns De La Fuente-Mella, Joaquin V. Fariña, Guilherme M. Sales

Abstract:

The primary approach for estimating bridge deterioration uses Markov-chain models and regression analysis. Traditional Markov models have problems in estimating the required transition probabilities when a small sample size is used. Often, reliable bridge data have not been taken over large periods, thus large data sets may not be available. This study presents an important change to the traditional approach by using the Small Data Method to estimate transition probabilities. The results illustrate that the Small Data Method and traditional approach both provide similar estimates; however, the former method provides results that are more conservative. That is, Small Data Method provided slightly lower than expected bridge condition ratings compared with the traditional approach. Considering that bridges are critical infrastructures, the Small Data Method, which uses more information and provides more conservative estimates, may be more appropriate when the available sample size is small. In addition, regression analysis was used to calculate bridge deterioration. Condition ratings were determined for bridge groups, and the best regression model was selected for each group. The results obtained were very similar to those obtained when using Markov chains; however, it is desirable to use more data for better results.

Keywords: concrete bridges, deterioration, Markov chains, probability matrix

Procedia PDF Downloads 338

Authors: Ali Al Moussaoui

Abstract:

An experiment was designed to assess the performance of 24 Lebanese adults (mean age 29:5 years) in a lexical decision making (LDM) task to find out how the facilitatory effect of phonological priming (PP) affects the speed of visual word recognition in Arabic as lexicality (wordhood) and phonological overlap positions (POP) vary. The experiment falls in line with previous research on phonological priming in the light of the cohort theory and in relation to visual word recognition. The experiment also departs from the research on the Arabic language in which the importance of the consonantal root as a distinct morphological unit is confirmed. Based on previous research, it is hypothesized that (1) PP has a facilitating effect in LDM with words but not with nonwords and (2) final phonological overlap between the prime and the target is more facilitatory than initial overlap. An LDM task was programmed on PsychoPy application. Participants had to decide if a target (e.g., bayn ‘between’) preceded by a prime (e.g., bayt ‘house’) is a word or not. There were 4 conditions: no PP (NP), nonwords priming nonwords (NN), nonwords priming words (NW), and words priming words (WW). The conditions were simultaneously controlled for word length, wordhood, and POP. The interstimulus interval was 700 ms. Within the PP conditions, POP was controlled for in which there were 3 overlap positions between the primes and the targets: initial (e.g., asad ‘lion’ and asaf ‘sorrow’), final (e.g., kattab ‘cause to write’ 2sg-mas and rattab ‘organize’ 2sg-mas), or two-segmented (e.g., namle ‘ant’ and naħle ‘bee’). There were 96 trials, 24 in each condition, using a within-subject design. The results show that concerning (1), the highest average reaction time (RT) is that in NN, followed firstly by NW and finally by WW. There is statistical significance only between the pairs NN-NW and NN-WW. Regarding (2), the shortest RT is that in the two-segmented overlap condition, followed by the final POP in the first place and the initial POP in the last place. The difference between the two-segmented and the initial overlap is significant, while other pairwise comparisons are not. Based on these results, PP emerges as a facilitatory phenomenon that is highly sensitive to lexicality and POP. While PP can have a facilitating effect under lexicality, it shows no facilitation in its absence, which intersects with several previous findings. Participants are found to be more sensitive to the final phonological overlap than the initial overlap, which also coincides with a body of earlier literature. The results contradict the cohort theory’s stress on the onset overlap position and, instead, give more weight to final overlap, and even heavier weight to the two-segmented one. In conclusion, this study confirms the facilitating effect of PP with words but not when stimuli (at least the primes and at most both the primes and targets) are nonwords. It also shows that the two-segmented priming is the most influential in LDM in Arabic.

Keywords: lexicality, phonological overlap positions, phonological priming, visual word recognition

Procedia PDF Downloads 189

Authors: Fan Ye

Abstract:

Adverse weather conditions, particularly those with low visibility, are critical to the driving tasks. However, the direct relationship between visibility distances and traffic flow/roadway safety is uncertain due to the limitation of visibility data availability. The recent growth of deployment of Road Weather Information Systems (RWIS) makes segment-specific visibility information available which can be integrated with other Intelligent Transportation System, such as automated warning system and variable speed limit, to improve mobility and safety. Before applying the RWIS visibility measurements in traffic study and operations, it is critical to validate the data. Therefore, an attempt was made in the paper to examine the validity and viability of RWIS visibility data by comparing visibility measurements among RWIS, airport weather stations, and weather information recorded by police in crash reports, based on Ohio data. The results indicated that RWIS visibility measurements were significantly different from airport visibility data in Ohio, but no conclusion regarding the reliability of RWIS visibility could be drawn in the consideration of no verified ground truth in the comparisons. It was suggested that more objective methods are needed to validate the RWIS visibility measurements, such as continuous in-field measurements associated with various weather events using calibrated visibility sensors.

Keywords: RWIS, visibility distance, low visibility, adverse weather

Procedia PDF Downloads 254

Authors: Rahul Malhotra

Abstract:

Fiber based access networks can deliver performance that can support the increasing demands for high speed connections. One of the new technologies that have emerged in recent years is Passive Optical Networks. This paper is targeted to show the simultaneous delivery of triple play service (data, voice and video). The comparative investigation and suitability of various data rates is presented. It is demonstrated that as we increase the data rate, number of users to be accommodated decreases due to increase in bit error rate.

Keywords: BER, PON, TDMPON, GPON, CWDM, OLT, ONT

Procedia PDF Downloads 561

Authors: Vahid Bayrami Rad

Abstract:

In this research, common methods and techniques have been investigated with a focus on intelligent fault finding and monitoring systems in the oil industry. In fact, remote and intelligent control methods are considered a necessity for implementing various operations in the oil industry, but benefiting from the knowledge extracted from countless data generated with the help of data mining algorithms. It is a avoid way to speed up the operational process for monitoring and troubleshooting in today's big oil companies. Therefore, by comparing data mining algorithms and checking the efficiency and structure and how these algorithms respond in different conditions, The proposed (Bayesian) algorithm using data clustering and their analysis and data evaluation using a colored Petri net has provided an applicable and dynamic model from the point of view of reliability and response time. Therefore, by using this method, it is possible to achieve a dynamic and consistent model of the remote control system and prevent the occurrence of leakage in oil pipelines and refineries and reduce costs and human and financial errors. Statistical data The data obtained from the evaluation process shows an increase in reliability, availability and high speed compared to other previous methods in this proposed method.

Keywords: wireless sensors, petroleum equipment troubleshooting, Bayesian algorithm, colored Petri net, rapid miner, data mining-reliability

Procedia PDF Downloads 72

Authors: T. Wen, Y. Liu, J. Wang, W. Zheng, T. Shao

Abstract:

Along with the increasing density of urban population, solar energy potential of building facade in high-density residential areas become a question that needs to be addressed. This paper studies how the solar energy utilization potential of building facades in different locations of a residential areas changes with different building layouts and orientations in Xining, a typical city in west China which possesses large solar radiation resource. Solar energy potential of three typical building layouts of residential areas, which are parallel determinant, gable misalignment, transverse misalignment, are discussed in detail. First of all, through the data collection and statistics of Xining new residential area, the most representative building parameters are extracted, including building layout, building height, building layers, and building shape. Secondly, according to the results of building parameters extraction, a general model is established and analyzed with rhinoceros 6.0 and its own plug-in grasshopper. Finally, results of the various simulations and data analyses are presented in a visualized way. The results show that there are great differences in the solar energy potential of building facades in different locations of residential areas under three typical building layouts. Generally speaking, the solar energy potential of the west peripheral location is the largest, followed by the East peripheral location, and the middle location is the smallest. When the deflection angle is the same, the solar energy potential shows the result that the West deflection is greater than the East deflection. In addition, the optimal building azimuth range under these three typical building layouts is obtained. Within this range, the solar energy potential of the residential area can always maintain a high level. Beyond this range, the solar energy potential drops sharply. Finally, it is found that when the solar energy potential is maximum, the deflection angle is not positive south, but 5 °or 15°south by west. The results of this study can provide decision analysis basis for residential design of Xining city to improve solar energy utilization potential and provide a reference for solar energy utilization design of urban residential buildings in other similar areas.

Keywords: building facade, solar energy potential, solar radiation, urban residential area, visualization, Xining city

Procedia PDF Downloads 183

Authors: Yasin Kutuk, Bengi Yanik Ilhan

Abstract:

Previous studies investigate the wage differentiations among regions in Turkey between couples who work in the same industry and those who work in different industries by using the models that is appropriate for cross sectional data. However, since there is no available panel data for this investigation in Turkey, pseudo panels using repeated cross-section data sets of the Household Labor Force Surveys 2004-2014 are employed in order to open a new way to examine wage differentiation patterns. For this purpose, household heads are separated into groups with respect to their household composition. These groups’ membership is assumed to be fixed over time such as age groups, education, gender, and NUTS1 (12 regions) Level. The average behavior of them can be tracked overtime same as in the panel data. Estimates using the pseudo panel data would be consistent with the estimates using genuine panel data on individuals if samples are representative of the population which has fixed composition, characteristics. With controlling the socioeconomic factors, wage differentiation of household income is affected by social, cultural and economic changes after global economic crisis emerged in US. It is also revealed whether wage differentiation is changing among the birth cohorts.

Keywords: wage income, same industry, pseudo panel, panel data econometrics

Procedia PDF Downloads 400

Authors: Kunal Shah, Swati Patel

Abstract:

Many real world problems involve data which can be considered as multi-label data streams. Efficient methods exist for multi-label classification in non streaming scenarios. However, learning in evolving streaming scenarios is more challenging, as the learners must be able to adapt to change using limited time and memory. Classification is used to predict class of unseen instance as accurate as possible. Multi label classification is a variant of single label classification where set of labels associated with single instance. Multi label classification is used by modern applications, such as text classification, functional genomics, image classification, music categorization etc. This paper introduces the task of multi-label classification, methods for multi-label classification and evolution measure for multi-label classification. Also, comparative analysis of multi label classification methods on the basis of theoretical study, and then on the basis of simulation was done on various data sets.

Keywords: binary relevance, concept drift, data stream mining, MLSC, multiple window with buffer

Procedia PDF Downloads 591

Authors: Anne-Marie Dionne

Abstract:

Teacher reading aloud to students is a practice that is well established in preschool and primary classrooms. Many benefits of this pedagogical activity have been highlighted in multiple studies. However, it has also been shown that teachers are not keen on choosing informational texts for their read aloud, as their selections for this venue are mainly fictional stories, mostly written in a unique narrative story-like structure. Considering that students soon have to read complex informational texts by themselves as they go from one grade to another, there is cause for concern because those who do not benefit from an early exposure to informational texts could be lacking knowledge of informational text structures that they will encounter regularly in their reading. Exposing students to informational texts could be done in different ways in classrooms. However, since read aloud appears to be such a common and efficient practice in preschool and primary grades, it is important to examine more deeply the factors taken into account by teachers when they are selecting their readings for this important teaching activity. Moreover, it seems critical to know why teachers are not inclined to choose more often informational texts when they are reading aloud to their pupils. A group of 22 preschool or primary grade teachers participated in this study. The data collection was done by a survey and an individual semi-structured interview. The survey was conducted in order to get quantitative data on the read-aloud practices of teachers. As for the interviews, they were organized around three categories of questions (exploratory, analytical, opinion) regarding the process of selecting the texts for the read-aloud sessions. A statistical analysis was conducted on the data obtained by the survey. As for the interviews, they were subjected to a content analysis aiming to classify the information collected in predetermined categories such as the reasons given to favor fictional texts over informative texts, the reasons given for avoiding informative texts for reading aloud, the perceptions of the challenges that the informative texts could bring when they are read aloud to students, and the perceived advantages that they would present if they were chosen more often for this activity. Results are showing variable factors that are guiding the teachers when they are making their selection of the texts to be read aloud. As for example, some of them are choosing solely fictional texts because of their convictions that these are more interesting for their students. They also perceive that the informational texts are not good choices because they are not suitable for pleasure reading. In that matter, results are pointing to some interesting elements. Many teachers perceive that read aloud of fictional or informational texts have different goals: fictional texts are read for pleasure and informational texts are read mostly for academic purposes. These results bring out the urgency for teachers to become aware of the numerous benefits that the reading aloud of each type of texts could bring to their students, especially the informational texts. The possible consequences of teachers’ perceptions will be discussed further in our presentation.

Keywords: fictional texts, informational texts, preschool or primary grade teachers, reading aloud

Procedia PDF Downloads 152

Authors: Kerem Ok, Serafettin Senturk, Serdar Aktas, Cem Cevikbas

Abstract:

Mobile technology is very popular nowadays and it provides a digital world where users can experience many value-added services. Service Providers are also eager to offer diverse value-added services to users such as digital identity, mobile financial services and so on. In this context, the security of data storage in smartphones and the security of communication between the smartphone and service provider are critical for the success of these services. In order to provide the required security functions, the SIM card is one acceptable alternative. Since SIM cards include a Secure Element, they are able to store sensitive data, create cryptographically secure keys, encrypt and decrypt data. In this paper, we design and implement a SIM and a smartphone framework that uses a SIM card for secure key generation, key storage, data encryption, data decryption and digital signing for mobile financial services. Our frameworks show that the SIM card can be used as a controlled Secure Element to provide required security functions for popular e-services such as mobile financial services.

Keywords: SIM card, mobile financial services, cryptography, secure data storage

Procedia PDF Downloads 316

Authors: Nastaran Khoshsabk

Abstract:

The evolution of communication through the Internet has influenced shaping and reshaping the self-presentation of social media users. Online communities both connect people and give voice to the voiceless allowing them to present themselves nationally and globally. People all around the world are experiencing censorship in different aspects of their life. Censorship can be externally imposed because of the political situations, or it can be self-imposed. Social media users choose the content they want to share and decide about the online audiences with whom they want to share this content. Most social media networks, such as Facebook, enable their users to be selective about the shared content and its availability to other people. However, sometimes instead of targeting a specific audience, users self-censor themselves or decide not to share various forms of information. These decisions are of particular importance in countries such as Iran where Internet is not the arena of free self-presentation and people are encouraged to stay away from political participation in the country and acting against the Islamic values. Facebook and some other social media tools are blocked in countries such as Iran. This project investigates the importance of social media in the life of Iranians to explore how they present themselves and construct their digital selves. The notion of cultural identity is applied in this research to explore the educational and informative role of social media in the identity formation and cultural representation of Facebook users. This study explores the self-censorship of Iranian adult Facebook users through their online self-representation and communication on the Internet. The data in this qualitative multiple case study have been collected through individual synchronous online interviews with the researcher’s Facebook friends and through the analysis of the participants’ Facebook profiles and activities over a period of six months. The data is analysed with an emphasis on the identity formation of participants through the recognition of the underlying themes. The exploration of online interviews is on the basis of participants’ personal accounts of self-censorship and cultural understanding through using social media. The driven codes and themes have been categorised considering censorship and place of culture on representation of self. Participants were asked to explain their views about censorship and conservatism through using social media. They reported their thoughts about deciding which content to share on Facebook and which to self-censor and their reasons behind these decisions. The codes and themes have been categorised considering censorship and its role in representation of idealised self. The ‘actual self’ showed to be hidden by an individual for different reasons such as its influence on their social status, academic achievements and job opportunities. It is hoped that this research will have implications for education contexts in countries that are experiencing social media filtering by offering an increased understanding of the importance of online communities; which can provide an educational environment to talk and learn about social taboos and constructing adults’ identity in virtual environment and through cultural self-presentation.

Keywords: cultural identity, identity formation, online communities, self-censorship

Procedia PDF Downloads 242

Authors: M. D. Antoine

Abstract:

Despite the widespread use of complementary and alternative medicine (CAM) for autistic children, little is known about the communication flow between the different parties involved in autism care (e.g., parents/caregivers, conventional providers, alternative practitioners). This study aimed to describe how communication occurs through the first year following an autism spectrum disorder (ASD) diagnosis to identify challenges and potential barriers to communication within the healthcare system in Ottawa, Canada. From an ecological perspective, we collected qualitative data through 12 semi-structured interviews with six parents/caregivers, three conventional providers (e.g., family doctor, neurodevelopmental pediatrician, psychologist), and three alternative practitioners (e.g., naturopath, occupational therapist, speech and language pathologist) operating in Ottawa. We interpreted the data using thematic analysis. Findings revealed communication challenges between the parents/caregivers and conventional providers while they experience better communication flow with fewer challenges in alternative care settings. However, parents/caregivers are the only links between the health professionals of both streams. From the five contexts examined: organizational, interpersonal, media, cultural, and political-legal, we found four themes (provider knowledge, care integration, flexible care, and time constraints) underlining specific barriers to communication flow between the parties involved in the care of autistic children. The increasing interest in alternative medicine is forcing changes in the healthcare system. Communications occur outside the norms making openings for better communication and information-sharing increasingly essential. Within the identified themes in the current study, the necessity for better communication between all parties involved in the care of autistic children is evident. More ASD and CAM-related training for providers would support effective parent/caregiver-provider communication. The findings of the current study contribute to a better understanding of the role of communication in the care management of autism, which has implications for effective autism care.

Keywords: alternative medicine, autism care management, autism spectrum disorder, conventional medicine, parent-provider communication

Procedia PDF Downloads 183

Authors: Rodlyn Remina Hines

Abstract:

Males and Females respond differently to life situations, including transnational living. Being in a transnational marriage relationship may put a strain on the relationship requiring partners to adjust their behaviors and expectancies of the other partner to accommodate the disruptions in the relationship. More so, when one partner is an immigrant to a new geographic location with the other in the native country, these disruptions may be intensive. This qualitative study examined gender differences in how married Ghanaian couples respond to making a life together as a couple while living across international borders. The study asked two questions: (1) What are the perceptions of males and females on transnational living? and (2) how do married males and females respond to transnational living situations? To answer these questions, semi-structured interviews were conducted with 24 married couples- with one partner living in the United States (U.S.) and the other spouse in Ghana via purposive and snowball sampling techniques. Participants were aged 26 to 59 years with an average age of 40; the average age of relationship: 10.41; and average years of living apart: 6.7. Induction and deduction hybrid analysis strategies were used to derive emerging themes. The results highlight significant gender differences in response to transnational living status and practices. The data indicate that transnational couples with the male spouse residing in the U.S. experience more relationship strains than is the case when the female partner is the immigrant. Three couples who were in divorce proceedings at the time of the interview had the male partner residing in the U.S. and the female spouse in Ghana. These gender differences also reflected spousal visitation frequency, duration of spousal reunification, amount of and frequency of spousal remittance(s), and immigration processing procedures. Finally, the data show female immigrant partners as better managers of transnational living stresses and strains than their male counterparts. Findings from this study have implications for marriage and family practitioners and immigration policy makers.

Keywords: gender differences, , ghanaian couples, ghanaian immigrants, transnational living

Procedia PDF Downloads 87

Authors: Samuel Oye Bandele

Abstract:

A policy is a system of principles to guide activities and strategic decisions of an organisation in order to achieve stated objectives and meeting expected outcomes. A Computer Based Test (CBT) policy is therefore a statement of intent to drive the CBT programmes, and should be implemented as a procedure or protocol. Policies are hence generally adopted by an organization or a nation. The concern here, in this paper, is the consideration and analysis of issues that are significant to evolving the acceptable policy that will drive the new CBT innovation in Nigeria. Public examinations and internal examinations in higher educational institutions in Nigeria are gradually making a radical shift from Paper Based or Paper-Pencil to Computer-Based Testing. The need to make an objective and empirical analysis of Policy issues relating to CBT became expedient. The following are some of the issues on CBT evolution in Nigeria that were identified as requiring policy backing. Prominent among them are requirements for establishing CBT centres, purpose of CBT, types and acquisition of CBT equipment, qualifications of staff: professional, technical and regular, security plans and curbing of cheating during examinations, among others. The descriptive research design was employed based on a population consisting of Principal Officers (Policymakers), Staff (Teaching and non-Teaching-Policy implementors), and CBT staff ( Technical and Professional- Policy supports) and candidates (internal and external). A fifty-item researcher-constructed questionnaire on policy issues was employed to collect data from 600 subjects drawn from higher institutions in South West Nigeria, using the purposive and stratified random sampling techniques. Data collected were analysed using descriptive (frequency counts, means and standard deviation) and inferential (t-test, ANOVA, regression and Factor analysis) techniques. Findings from this study showed, among others, that the factor loadings had significantly weights on the organizational and National policy issues on CBT innovation in Nigeria.

Keywords: computer-based testing, examination, innovation, paper-based testing, paper pencil based testing, policy issues

Procedia PDF Downloads 251

Authors: Salma El Hajjami, Jamal Malki, Alain Bouju, Mohammed Berrada

Abstract:

With the expansion of machine learning and data mining in the context of Big Data analytics, the common problem that affects data is class imbalance. It refers to an imbalanced distribution of instances belonging to each class. This problem is present in many real world applications such as fraud detection, network intrusion detection, medical diagnostics, etc. In these cases, data instances labeled negatively are significantly more numerous than the instances labeled positively. When this difference is too large, the learning system may face difficulty when tackling this problem, since it is initially designed to work in relatively balanced class distribution scenarios. Another important problem, which usually accompanies these imbalanced data, is the overlapping instances between the two classes. It is commonly referred to as noise or overlapping data. In this article, we propose an approach called: One Side Behavioral Noise Reduction (OSBNR). This approach presents a way to deal with the problem of class imbalance in the presence of a high noise level. OSBNR is based on two steps. Firstly, a cluster analysis is applied to groups similar instances from the minority class into several behavior clusters. Secondly, we select and eliminate the instances of the majority class, considered as behavioral noise, which overlap with behavior clusters of the minority class. The results of experiments carried out on a representative public dataset confirm that the proposed approach is efficient for the treatment of class imbalances in the presence of noise.

Keywords: machine learning, imbalanced data, data mining, big data

Procedia PDF Downloads 134

Authors: Areej Salaymeh, Loren Schwiebert, Stephen Remias, Jonathan Waddell

Abstract:

Extracting information from new data sources has emerged as a crucial task in many traffic planning processes, such as identifying traffic patterns, route planning, traffic forecasting, and locating infrastructure improvements. Given the advanced technologies used to collect Global Positioning System (GPS) data from dedicated GPS devices, GPS equipped phones, and navigation tools, intelligent data analysis methodologies are necessary to mine this raw data. In this research, an automatic detection framework is proposed to help identify and classify the locations of stopped GPS waypoints into two main categories: signalized intersections or highway congestion. The Delaunay triangulation is used to perform this assessment in the clustering phase. While most of the existing clustering algorithms need assumptions about the data distribution, the effectiveness of the Delaunay triangulation relies on triangulating geographical data points without such assumptions. Our proposed method starts by cleaning noise from the data and normalizing it. Next, the framework will identify stoppage points by calculating the traveled distance. The last step is to use clustering to form groups of waypoints for signalized traffic and highway congestion. Next, a binary classifier was applied to find distinguish highway congestion from signalized stop points. The binary classifier uses the length of the cluster to find congestion. The proposed framework shows high accuracy for identifying the stop positions and congestion points in around 99.2% of trials. We show that it is possible, using limited GPS data, to distinguish with high accuracy.

Keywords: Delaunay triangulation, clustering, intelligent transportation systems, GPS data

Procedia PDF Downloads 280

Authors: Pedro A. Marques, Francisco Monteiro, Alessandra Zumbo, Alessia Simonini, Miguel A. Mendez

Abstract:

Cryogenic fuels such as Liquid Hydrogen (LH₂) must be transported and stored at extremely low temperatures. Without expensive active cooling solutions, preventing fuel boil-off over time is impossible. Hence, one must resort to venting systems at the cost of significant energy and fuel mass loss. These losses increase significantly in propellant tanks installed on vehicles, as the presence of external accelerations induces sloshing. Sloshing increases heat and mass transfer rates and leads to significant pressure oscillations, which might further trigger propellant venting. To make LH₂ economically viable, it is essential to minimize these factors by using advanced control techniques. However, these require accurate modelling and a full understanding of the tank's thermodynamics. The present research aims to implement a simple thermodynamic model capable of predicting the state of a cryogenic fuel tank under different operating conditions (i.e., filling, pressurization, fuel extraction, long-term storage, and sloshing). Since this model relies on a set of closure parameters to drive the system's transient response, it must be calibrated using experimental or numerical data. This work focuses on the former approach, wherein the model is calibrated through an experimental campaign carried out on a reduced-scale model of a cryogenic tank. The thermodynamic model of the system is composed of three control volumes: the ullage, the liquid, and the insulating walls. Under this lumped formulation, the governing equations are derived from energy and mass balances in each region, with mass-averaged properties assigned to each of them. The gas-liquid interface is treated as an infinitesimally thin region across which both phases can exchange mass and heat. This results in a coupled system of ordinary differential equations, which must be closed with heat and mass transfer coefficients between each control volume. These parameters are linked to the system evolution via empirical relations derived from different operating regimes of the tank. The derivation of these relations is carried out using an inverse method to find the optimal relations that allow the model to reproduce the available data. This approach extends classic system identification methods beyond linear dynamical systems via a nonlinear optimization step. Thanks to the data-driven assimilation of the closure problem, the resulting model accurately predicts the evolution of the tank's thermodynamics at a negligible computational cost. The lumped model can thus be easily integrated with other submodels to perform complete system simulations in real time. Moreover, by setting the model in a dimensionless form, a scaling analysis allowed us to relate the tested configurations to a representative full-size tank for naval applications. It was thus possible to compare the relative importance of different transport phenomena between the laboratory model and the full-size prototype among the different operating regimes.

Keywords: destratification, hydrogen, modeling, pressure-drop, pressurization, sloshing, thermodynamics

Procedia PDF Downloads 98

Authors: Gayathri Nagarajan, L. D. Dhinesh Babu

Abstract:

Health care is one of the prominent industries that generate voluminous data thereby finding the need of machine learning techniques with big data solutions for efficient processing and prediction. Missing data, incomplete data, real time streaming data, sensitive data, privacy, heterogeneity are few of the common challenges to be addressed for efficient processing and mining of health care data. In comparison with other applications, accuracy and fast processing are of higher importance for health care applications as they are related to the human life directly. Though there are many machine learning techniques and big data solutions used for efficient processing and prediction in health care data, different techniques and different frameworks are proved to be effective for different applications largely depending on the characteristics of the datasets. In this paper, we present a framework that uses ensemble machine learning technique gradient boosted trees for data classification in health care big data. The framework is built on Spark platform which is fast in comparison with other traditional frameworks. Unlike other works that focus on a single technique, our work presents a comparison of six different machine learning techniques along with gradient boosted trees on datasets of different characteristics. Five benchmark health care datasets are considered for experimentation, and the results of different machine learning techniques are discussed in comparison with gradient boosted trees. The metric chosen for comparison is misclassification error rate and the run time of the algorithms. The goal of this paper is to i) Compare the performance of gradient boosted trees with other machine learning techniques in Spark platform specifically for health care big data and ii) Discuss the results from the experiments conducted on datasets of different characteristics thereby drawing inference and conclusion. The experimental results show that the accuracy is largely dependent on the characteristics of the datasets for other machine learning techniques whereas gradient boosting trees yields reasonably stable results in terms of accuracy without largely depending on the dataset characteristics.

Keywords: big data analytics, ensemble machine learning, gradient boosted trees, Spark platform

Procedia PDF Downloads 244

Authors: Raahil Sheikh, Abhishek Maurya, Priya Gujjar, Himanshu Dwivedi, Prathamesh Minde

Abstract:

UAVs have been an initial member of our environment since it's the first used by Austrian warfare in Venice. At that stage, they were just pilotless balloons equipped with bombs to be dropped on enemy territory. Over time, technological advancements allowed UAVs to be controlled remotely or autonomously. This study shall mainly focus on the intensification of pre-existing manual drones equipping them with a variety of sensors and making them autonomous, and capable, and purposing them for a variety of roles, including thermal sensing, data collection, tracking creatures, forest fires, volcano detection, hydrothermal studies, urban heat, Island measurement, and other environmental research. The system can also be used for reconnaissance, research, 3D mapping, and search and rescue missions. This study mainly focuses on automating tedious tasks and reducing human errors as much as possible, reducing deployment time, and increasing the overall efficiency, efficacy, and reliability of the UAVs. Creation of a comprehensive Ground Control System UI (GCS) enabling less trained professionals to be able to use the UAV with maximum potency. With the inclusion of such an autonomous system, artificially intelligent paths and environmental gusts and concerns can be avoided.

Keywords: UAV, drone, autonomous system, thermal imaging

Procedia PDF Downloads 79

Authors: Michelann Parr

Abstract:

Family-school-community engagement enhances family-school-community well-being, collective confidence, and school climate. While it is often referred to as a positive thing in the literature for families, schools, and communities, it does not come without its struggles. While there are numerous things families, schools, and communities do each and every day to enhance engagement, it is often difficult to find our way to true belonging and engagement. Working our way surface level barriers is easy; we can provide childcare, transportation, resources, and refreshments. We can even change the environment so that families will feel welcome, valued, and respected. But there are often mindsets and perpsectives buried deep below the surface, most often grounded in societal, familial, and political norms, expectations, pressures, and narratives. This work requires ongoing energy, commitment, and engagement of all stakeholders, including families, schools, and communities. Each and every day, we need to take a reflective and introspective stance at what is said and done and how it supports the overall goal of family-school-community engagement. And whatever we must occur within a paradigm of care in additional to one of critical thinking and social justice. Families, and those working with families, must not simply accept all that is given, but should instead ask these types of questions: a) How, and by whom, are the current philosophies and practices of family-school engagement interrogated? b) How might digging below surface level meanings support understanding of what is being said and done? c) How can we move toward meaningful and authentic engagement that balances knowledge and power between family, school, district, community (local and global), and government? This type of work requires conscious attention and intentional decision-making at all levels bringing us one step closer to authentic and meaningful partnerships. Strategies useful to building a growth mindset include: a) interrogating and exploring consistencies and inconsistencies by looking at what is done and what is not done through multiple perspectives; b) recognizing that enhancing family-engagement and changing mindsets take place at the micro-level (e.g., family and school), but also require active engagement and awareness at the macro-level (e.g., community agencies, district school boards, government); c) taking action as an advocate or activist. Negative narratives about families, schools, and communities should not be maintained, but instead critical and courageous conversations in and out of school should be initiated and sustained; and d) maintaining consistency, simplicity, and steady progress. All involved in engagement need to be aware of the struggles, but keep them in check with the many successes. Change may not be observed on a day-to-day basis or even immediately, but stepping back and looking from the outside in, might change the view. Working toward a growth mindset will produce better results than a fixed mindset, and this takes time.

Keywords: family engagment, family-school-community engagement, parent engagement, parent involvment

Procedia PDF Downloads 185

Authors: Joe Collins, Matthias Rother

Abstract:

Background: Oral non-steroidal anti-inflammatory drugs (NSAIDs) are widely used for the relief of joint pain during and post-exercise. However, oral NSAIDs increase the risk of systemic side effects, even in healthy individuals, and retard recovery from muscle soreness. TDT 064 (Flexiseq®), a topical formulation containing ultra-deformable drug-free Sequessome™ vesicles, has demonstrated equivalent efficacy to oral celecoxib in reducing osteoarthritis-associated joint pain and stiffness. TDT 064 does not cause NSAID-related adverse effects. We describe clinical study data and a case report on the effectiveness of TDT 064 in reducing joint pain after exercise. Methods: Participants with a pain score ≥3 (10-point scale) 12–16 hours post-exercise were randomized to receive TDT 064 plus oral placebo, TDT 064 plus oral ketoprofen, or ketoprofen in ultra-deformable phospholipid vesicles plus oral placebo. Results: In the 168 study participants, pain scores were significantly higher with oral ketoprofen plus TDT 064 than with TDT 064 plus placebo in the 7 days post-exercise (P = 0.0240) and recovery from muscle soreness was significantly longer (P = 0.0262). There was a low incidence of adverse events. These data are supported by clinical experience. A 24-year-old male professional rugby player suffered a traumatic lisfranc fracture in March 2014 and underwent operative reconstruction. He had no relevant medical history and was not receiving concomitant medications. He had undergone anterior cruciate ligament reconstruction in 2008. The patient reported restricted training due to pain (score 7/10), stiffness (score 9/10) and poor function, as well as pain when changing direction and running on consecutive days. In July 2014 he started using TDT 064 twice daily at the recommended dose. In November 2014 he noted reduced pain on running (score 2-3/10), decreased morning stiffness (score 4/10) and improved joint mobility and was able to return to competitive rugby without restrictions. No side effects of TDT 064 were reported. Conclusions: TDT 064 shows efficacy against exercise- and injury-induced joint pain, as well as that associated with osteoarthritis. It does not retard muscle soreness recovery after exercise compared with an oral NSAID, making it an alternative approach for the treatment of joint pain during and post-exercise.

Keywords: exercise, joint pain, TDT 064, phospholipid vesicles

Procedia PDF Downloads 480

Authors: Joao Victor Borges Dos Santos, Thomas Richard, Yevhen Kovalyshen

Abstract:

Down-the-hole (DTH) percussive drilling is an excavation method that is widely used in the mining industry due to its high efficiency in fragmenting hard rock formations. A DTH hammer system consists of a fluid driven (air or water) piston and a drill bit; the reciprocating movement of the piston transmits its kinetic energy to the drill bit by means of stress waves that propagate through the drill bit towards the rock formation. In the literature of percussive drilling, the existence of an optimum drilling state (Sweet Spot) is reported in some laboratory and field experimental studies. An optimum rate of penetration is achieved for a specific range of axial thrust (or weight-on-bit) beyond which the rate of penetration decreases. Several authors advance different explanations as possible root causes to the occurrence of the Sweet Spot, but a universal explanation or consensus does not exist yet. The experimental investigation in this work was initiated with drilling experiments conducted at a mining site. A full-scale drilling rig (equipped with a DTH hammer system) was instrumented with high precision sensors sampled at a very high sampling rate (kHz). Data was collected while two boreholes were being excavated, an in depth analysis of the recorded data confirmed that an optimum performance can be achieved for specific ranges of input thrust (weight-on-bit). The high sampling rate allowed to identify the bit penetration at each single impact (of the piston on the drill bit) as well as the impact frequency. These measurements provide a direct method to identify when the hammer does not fire, and drilling occurs without percussion, and the bit propagate the borehole by shearing the rock. The second stage of the experimental investigation was conducted in a laboratory environment with a custom-built equipment dubbed Woody. Woody allows the drilling of shallow holes few centimetres deep by successive discrete impacts from a piston. After each individual impact, the bit angular position is incremented by a fixed amount, the piston is moved back to its initial position at the top of the barrel, and the air pressure and thrust are set back to their pre-set values. The goal is to explore whether the observed optimum drilling state stems from the interaction between the drill bit and the rock (during impact) or governed by the overall system dynamics (between impacts). The experiments were conducted on samples of Calca Red, with a drill bit of 74 millimetres (outside diameter) and with weight-on-bit ranging from 0.3 kN to 3.7 kN. Results show that under the same piston impact energy and constant angular displacement of 15 degrees between impact, the average drill bit rate of penetration is independent of the weight-on-bit, which suggests that the sweet spot is not caused by intrinsic properties of the bit-rock interface.

Keywords: optimum drilling state, experimental investigation, field experiments, laboratory experiments, down-the-hole percussive drilling

Procedia PDF Downloads 96

Authors: Yori Gidron, Laura Caton, Irit Ben-Aharon

Abstract:

Background: Many patients and even certain clinicians attribute cancer development to psychosocial factors. Yet, empirical data supports more the prognostic role, rather than the etiological role, of psychosocial factors in cancer. Part of the inconsistency may result from not considering possible moderating factors in the etiological role of psychosocial factors. One important candidate moderating factor is the vagal nerve, whose activity is indexed by heart-rate variability (HRV). The vagal nerve may prevent cancer since it reduces inflammation on the one hand, and since it increases anti-tumor immunity on the other hand. This study examined the moderating role of the vagus in the relation between life threatening events (LTE) and cancer development. Method: We re-analyzed data from the Lifelines Dutch longitudinal cohort study of over 150,000 people. The present study included 82,751 adults, who initially were cancer-free. We extracted information on background factors (e.g., age, gender, fat consumption), whether they ever experienced LTE, HRV and cancer diagnosis as reported by patients in annual clinic visits. HRV was derived from brief ECGs. Results: Of the full sample, 1011 people developed cancer during a follow-up. In the full sample, LTE significantly predicted cancer development (R.R = 1.063 p < .01) and HRV significantly predicted a reduced risk of cancer development (R.R = .506 p <.001). Importantly, LTE significantly predicted cancer only when HRV was low (R.R = 1.056, 95% CI: 1.007 - 1.108, p < .05) but not when HRV was high (R.R = 1.014; 95% CI: 0.916 - 1.122, p > 0.05), independent of confounders. Conclusions: To the best of our knowledge, this is the first study showing in a large sample that LTE predict cancer development, and that this occurs only when vagal nerve activity (HRV) is relatively low. These results could result from lack of vagal modulation of inflammation and also from lack of vagal modulation of stress responses. Results are in line with the cancer-protective role of the vagus. HRV needs to be routinely monitored in the population and future intervention trials need to examine whether vagal nerve activation can prevent cancer in people with LTE and with other cancer risk factors.

Keywords: cancer development, life-events, moderation, vagal nerve

Procedia PDF Downloads 174

Authors: Kanika Gupta, Ashok Kumar

Abstract:

Biofilms are dense, highly hydrated cell clusters that are irreversibly attached to a substratum, to an interface or to each other, and are embedded in a self-produced gelatinous matrix composed of extracellular polymeric substances. Research in biofilm field has become very significant, as biofilm has shown high mechanical resilience and resistance to antibiotic treatment and constituted as a significant problem in both healthcare and other industry related to microorganisms. The massive information both stated and hidden in the biofilm literature are growing exponentially therefore it is not possible for researchers and practitioners to automatically extract and relate information from different written resources. So, the current work proposes and discusses the use of text mining techniques for the extraction of information from biofilm literature corpora containing 34306 documents. It is very difficult and expensive to obtain annotated material for biomedical literature as the literature is unstructured i.e. free-text. Therefore, we considered unsupervised approach, where no annotated training is necessary and using this approach we developed a system that will classify the text on the basis of growth and development, drug effects, radiation effects, classification and physiology of biofilms. For this, a two-step structure was used where the first step is to extract keywords from the biofilm literature using a metathesaurus and standard natural language processing tools like Rapid Miner_v5.3 and the second step is to discover relations between the genes extracted from the whole set of biofilm literature using pubmed.mineR_v1.0.11. We used unsupervised approach, which is the machine learning task of inferring a function to describe hidden structure from 'unlabeled' data, in the above-extracted datasets to develop classifiers using WinPython-64 bit_v3.5.4.0Qt5 and R studio_v0.99.467 packages which will automatically classify the text by using the mentioned sets. The developed classifiers were tested on a large data set of biofilm literature which showed that the unsupervised approach proposed is promising as well as suited for a semi-automatic labeling of the extracted relations. The entire information was stored in the relational database which was hosted locally on the server. The generated biofilm vocabulary and genes relations will be significant for researchers dealing with biofilm research, making their search easy and efficient as the keywords and genes could be directly mapped with the documents used for database development.

Keywords: biofilms literature, classifiers development, text mining, unsupervised learning approach, unstructured data, relational database

Procedia PDF Downloads 174

Authors: G. Carrero, C. Contreras, M. J. Hendzel

Abstract:

Linker histones or histones H1 are highly mobile nuclear proteins that regulate the organization of chromatin and limit DNA accessibility by binding to the chromatin structure (DNA and associated proteins). It is known that this binding process is driven by both slow (strong binding) and rapid (weak binding) interactions. However, the exact binding mechanism has not been fully described. Moreover, the existing models only account for one type of bound population that does not distinguish explicitly between the weakly and strongly bound proteins. Thus, we propose different systems of reaction-diffusion equations to describe explicitly the rapid and slow interactions during a FRAP (Fluorescence Recovery After Photobleaching) experiment. We perform a model comparison analysis to characterize the binding mechanism of histone H1 and provide new meaningful biophysical information on the kinetics of histone H1.

Keywords: FRAP (Fluorescence Recovery After Photobleaching), histone H1, histone H1 binding kinetics, linker histone, reaction-diffusion equation

Procedia PDF Downloads 445

Authors: Luke Taylor, Samantha J. Hall, Kenneth I. Cumming, Jakki Bardsley, Scott S. P. Wildman

Abstract:

Medway School of Pharmacy, as part of an Equality Diversity and Inclusivity (EDI) initiative run by the University of Kent, decided to take steps to try and negate disparities in numeracy competencies within students undertaking the Master of Pharmacy degree in order to combat a trend in pharmacy students’ numerical abilities upon entry. This included a research driven project 1) to identify if pharmacy students are aware of weaknesses in their numeracy capabilities, and 2) recognise where their numeracy skillset is lacking. In addition to gaining this student perspective, a number of actions have been implemented to support students in improving their numeracy competencies. Reflective and quantitative analysis has shown promising improvements for the final year cohort of 2014/15 when compared to previous years. The method of involving student feedback into the structure of numeracy teaching/support has proven to be extremely beneficial to both students and teaching staff alike. Students have felt empowered and in control of their own learning requirements, leading to increased engagement and attainment. School teaching staff have received quality data to help improve existing initiatives and to innovate further in the area of numeracy teaching. In light of the recognised improvements, further actions are currently being trialled in the area of numeracy support. This involves utilising Virtual Learning Environment platforms to provide individualised support as a supplement to the increased numeracy mentoring (staff and peer) provided to students. Mentors who provide group or one-to-one sessions are now given significant levels of training in dealing with situations that commonly arise from mentoring schemes. They are also provided with continued support throughout the life of their degree. Following results from this study, Medway School of Pharmacy hopes to drive increasing numeracy standards within Pharmacy (primarily through championing peer mentoring) as well as other healthcare professions including Midwifery and Nursing.

Keywords: attainment, ethnicity, numeracy, pharmacy, support

Procedia PDF Downloads 237

Authors: Afrin Jaman Bonny, Mehrin Jahan, Zannatul Ferdhoush, Mumenunnessa Keya, Md. Shihab Mahmud, Sharun Akter Khushbu, Sheak Rashed Haider Noori, Sheikh Abujar

Abstract:

In 2020, the SARS-CoV-2 pandemic had led all the educational institutions to move to online learning platforms to ensure safety as well as the continuation of learning without any disruption to students’ academic life. But after the reopening of those educational institutions suddenly in Bangladesh, it became a vital demand to observe students take on this decision and how much they are comfortable with the new habits. When all educational institutions were ordered to re-open after more than a year, data was collected from students of all educational levels. A Google Form was used to conduct this online survey, and a total of 565 students participated without being pressured. The survey reveals the students' preferences for online and offline education systems, as well as their mental health at the time including their behavior to get back to offline classes depending on getting vaccinated or not. After evaluating the findings, it is clear that respondents' choices vary depending on gender and educational level, with female and male participants experiencing various mental health difficulties and attitudes toward returning to offline classes. As a result of this study, the student’s overall perspective on the sudden reopening of their educational institutions has been analyzed.

Keywords: covid-19 epidemic, educational proceeding, university students, school/college students, physical activity, online platforms, mental health, psychological distress

Procedia PDF Downloads 217

Authors: Paul E. Priester

Abstract:

The Twelve Step approach to recovery from substance use and addictive disorders is considered an evidence-based model that assists many who recover from a chronic, progressive, fatal disease. Two key processes that contribute to the success of obtaining recovery from substance use disorders (SUD) are meeting engagement and the development of spiritual beliefs. Beyond establishing that there is a positive relationship between the development of spiritual beliefs in recovery from SUD’s, there has been a paucity of research exploring individual differences among individuals in this development of spiritual beliefs. One such personality variable that deserves exploration is that of the need for cognition. The need for cognition is a personality variable that explains the cognitive style of individuals. Individuals with a high need for cognition enjoy examining the complexities of a situation before coming to a conclusion. While individuals with a low need for cognition do not value or spend time cognitively dissecting a situation or decision. It is important to point out that a high need for cognition does not necessarily imply a high level of cognitive ability. Indeed, one could make the argument that a low need for cognition individual is not “wasting” cognitive energy in perseverating the multitude of aspects of a particular decision. This paper will present two case studies demonstrating the development of spiritual beliefs that enabled long-term recovery from SUD. The first case study presents an agnostic individual with a low need for cognition cognitive style in his development of spirituality in support of his recovery from alcoholism within the context of Alcoholics Anonymous. The second case study represents an adamant atheist with a high need for cognition cognitive style. This second individual is an intravenous cocaine addict and alcoholic who recovers through the development of spirituality within the contexts of Alcoholics Anonymous and Narcotics Anonymous. The two case studies will be contrasted with each other, noting how the individuals’ cognitive style mediated the development of spirituality that supported their long-term recovery from alcoholism and addiction.

Keywords: spirituality, twelve step recovery, need for cognition, individual differences in recovery from addictions

Procedia PDF Downloads 98