Search results for: imbalanced data with class overlap

Authors: Thrivikraman Aswathi, S. Advaith

Abstract:

As there can be cases where the use of real data is somehow limited, such as when it is hard to get access to a large volume of real data, we need to go for synthetic data generation. This produces high-quality synthetic data while maintaining the statistical properties of a specific dataset. In the present work, a generative adversarial network (GAN) is trained to produce multivariate time series (MTS) data since the MTS is now being gathered more often in various real-world systems. Furthermore, the GAN-generated MTS data is fed into a transformer-based deep learning architecture that carries out the data categorization into predefined classes. Further, the model is evaluated across various distinct domains by generating corresponding MTS data.

Keywords: GAN, transformer, classification, multivariate time series

Procedia PDF Downloads 123

Authors: Lakshmi Prayaga, Chandra Prayaga. Aaron Wade, Gopi Shankar Mallu, Harsha Satya Pola

Abstract:

Synthetic data generated by Generative Adversarial Networks and Autoencoders is becoming more common to combat the problem of insufficient data for research purposes. However, generating synthetic data is a tedious task requiring extensive mathematical and programming background. Open-source platforms such as the Synthetic Data Vault (SDV) and Mostly AI have offered a platform that is user-friendly and accessible to non-technical professionals to generate synthetic data to augment existing data for further analysis. The SDV also provides for additions to the generic GAN, such as the Gaussian copula. We present the results from two synthetic data sets (CTGAN data and CTGAN with Gaussian Copula) generated by the SDV and report the findings. The results indicate that the ROC and AUC curves for the data generated by adding the layer of Gaussian copula are much higher than the data generated by the CTGAN.

Keywords: synthetic data generation, generative adversarial networks, conditional tabular GAN, Gaussian copula

Procedia PDF Downloads 72

Authors: Habib Muhammetoglu, I. Ethem Karadirek, Selami Kara, Ayse Muhammetoglu

Abstract:

The problem of non-revenue water (NRW) from municipal water distribution networks is common in many countries such as Turkey, where the average yearly water losses are around 50% . Water losses can be divided into two major types namely: 1) Real or physical water losses, and 2) Apparent or commercial water losses. Total water losses in Antalya city, Turkey is around 45%. Methods: A research study was conducted to develop appropriate methodologies to reduce NRW. A pilot study area of about 60 thousands inhabitants was chosen to apply the study. The pilot study area has a supervisory control and data acquisition (SCADA) system for the monitoring and control of many water quantity and quality parameters at the groundwater drinking wells, pumping stations, distribution reservoirs, and along the water mains. The pilot study area was divided into 18 District Metered Areas (DMAs) with different number of service connections that ranged between a few connections to less than 3000 connections. The flow rate and water pressure to each DMA were on-line continuously measured by an accurate flow meter and water pressure meter that were connected to the SCADA system. Customer water meters were installed to all billed and unbilled water users. The monthly water consumption as given by the water meters were recorded regularly. Water balance was carried out for each DMA using the well-know standard IWA approach. There were considerable variations in the water losses percentages and the components of the water losses among the DMAs of the pilot study area. Old Class B customer water meters at one DMA were replaced by more accurate new Class C water meters. Hydraulic modelling using the US-EPA EPANET model was carried out in the pilot study area for the prediction of water pressure variations at each DMA. The data sets required to calibrate and verify the hydraulic model were supplied by the SCADA system. It was noticed that a number of the DMAs exhibited high water pressure values. Therefore, pressure reducing valves (PRV) with constant head were installed to reduce the pressure up to a suitable level that was determined by the hydraulic model. On the other hand, the hydraulic model revealed that the water pressure at the other DMAs cannot be reduced when complying with the minimum pressure requirement (3 bars) as stated by the related standards. Results: Physical water losses were reduced considerably as a result of just reducing water pressure. Further physical water losses reduction was achieved by applying acoustic methods. The results of the water balances helped in identifying the DMAs that have considerable physical losses. Many bursts were detected especially in the DMAs that have high physical water losses. The SCADA system was very useful to assess the efficiency level of this method and to check the quality of repairs. Regarding apparent water losses reduction, changing the customer water meters resulted in increasing water revenue by more than 20%. Conclusions: DMA, SCADA, modelling, pressure management, leakage detection and accurate customer water meters are efficient for NRW.

Keywords: NRW, water losses, pressure management, SCADA, apparent water losses, urban water distribution networks

Procedia PDF Downloads 392

Authors: Saeyoung Lee

Abstract:

As the enthusiasm for education in Korea is too much high, it is well known by others that children in Korea get good scores in Mathematics. However, behind of this good reputation, children in Korea are easy to get lose self-confidence, tend to complaint and rarely participate in the class because of too much competition which leads to lack of competences. In this regard, the main goals of this paper are, by applying the programme based on peer-communication on Mathematics education field, it would like to improve self-managemental competence to make children gain self-confidence, communicative competence to make them deal with complaint and communitive competence to make them participated in the class for the age of 10 children to solve this problem. 14 children the age of 10 in one primary school in Gangnam, Seoul, Korea had participated in the research from March 2018 to October 2018. They were under the programme based on peer-communication during the period. Every Mathematics class maintained the same way. Firstly a problem was given to children. Secondly, children were asked to find many ways to solve the problem as much as they could by themselves. Thirdly all ways to solve the problem by children were posted on the board and three of the children made a group to distinguish the ways from valid to invalid. Lastly, all children made a discuss to find one way which is the most efficient among valid ways. Pre-test was carried out by the questionnaire based on Likert scale before applying the programme. The result of the pre-test was 3.89 for self-managemental competence, 3.91 for communicative competence and 4.19 for communitive competence. Post-test was carried out by the same questionnaire after applying the programme. The result of the post-test was 3.93 for self-managemental competence, 4.23 for communicative competence and 4.20 for communitive competence. That means by applying the programme based on peer-communication on Mathematics education field, the age of 10 children in Korea could improve self-managemental, communicative and communitive competence. Especially it works very well on communicative competence by increasing 0.32 points as it marked. Considering this research, Korean Mathematics education based on competition which leads to lack of competences should be changed to cooperative structure to make students more competent rather than just getting good scores. In conclusion, innovative teaching methods which are focused on improving competences such as the programme based on peer-communication which was applied in this research are strongly required to be studied and widely used.

Keywords: competences, mathematics education, peer-communication, primary education

Procedia PDF Downloads 129

Authors: Keyao Li, Sai On Cheung

Abstract:

Incorporation of alternative dispute resolution (ADR) mechanism has been the main feature of current trend of construction project dispute resolution (CPDR). ADR approaches have been identified as efficient mechanisms and are suitable alternatives to litigation and arbitration. Moreover, the use of ADR in this multi-tiered dispute resolution process often leads to repeated evaluations of a same dispute. Multi-tiered CPDR may become a breeding ground for cognitive biases. When completed knowledge is not available at the early tier of construction dispute resolution, disputing parties may form preconception of the dispute matter or the counterpart. This preconception would influence their information processing in the subsequent tier. Disputing parties tend to search and interpret further information in a self-defensive way to confirm their early positions. Their imbalanced information collection would boost their confidence in the held assessments. Their attitudes would be hardened and difficult to compromise. The occurrence of cognitive bias, therefore, impedes efficient dispute settlement. This study aims to explore ways to minimize bias in CPDR. Based on a comprehensive literature review, three types of bias minimizing approaches were collected: strategy-based, attitude-based and process-based. These approaches were further operationalized into bias minimizing measures. To verify the usefulness and practicability of these bias minimizing measures, semi-structured interviews were conducted with ten CPDR third party neutral professionals. All of the interviewees have at least twenty years of experience in facilitating settlement of construction dispute. The usefulness, as well as the implications of the bias minimizing measures, were validated and suggested by these experts. There are few studies on cognitive bias in construction management in general and in CPDR in particular. This study would be the first of its type to enhance the efficiency of construction dispute resolution by highlighting strategies to minimize the biases therein.

Keywords: bias, construction project dispute resolution, minimization, multi-tiered, semi-structured interview

Procedia PDF Downloads 182

Authors: Feng Tao, Ma Jing, Guo Xian, Wang Jing

Abstract:

Named Data Networking (NDN) replaces IP address of traditional network with data name, and adopts dynamic cache mechanism. In the existing mechanism, however, only one-to-one search can be achieved because every data has a unique name corresponding to it. There is a certain mapping relationship between data content and data name, so if the data name is intercepted by an adversary, the privacy of the data content and user’s interest can hardly be guaranteed. In order to solve this problem, this paper proposes a one-to-many fuzzy search scheme based on order-preserving encryption to reduce the query overhead by optimizing the caching strategy. In this scheme, we use hash value to ensure the user’s query safe from each node in the process of search, so does the privacy of the requiring data content.

Keywords: NDN, order-preserving encryption, fuzzy search, privacy

Procedia PDF Downloads 478

Authors: Renukswamy Chikkamath, Vishvapalsinhji Ramsinh Parmar, Christoph Hewel, Markus Endres

Abstract:

Given a patent document, identifying distinct semantic annotations is an interesting research aspect. Text annotation helps the patent practitioners such as examiners and patent attorneys to quickly identify the key arguments of any invention, successively providing a timely marking of a patent text. In the process of manual patent analysis, to attain better readability, recognising the semantic information by marking paragraphs is in practice. This semantic annotation process is laborious and time-consuming. To alleviate such a problem, we proposed a dataset to train machine learning algorithms to automate the highlighting process. The contributions of this work are: i) we developed a multi-class dataset of size 150k samples by traversing USPTO patents over a decade, ii) articulated statistics and distributions of data using imperative exploratory data analysis, iii) baseline Machine Learning models are developed to utilize the dataset to address patent paragraph highlighting task, and iv) future path to extend this work using Deep Learning and domain-specific pre-trained language models to develop a tool to highlight is provided. This work assists patent practitioners in highlighting semantic information automatically and aids in creating a sustainable and efficient patent analysis using the aptitude of machine learning.

Keywords: machine learning, patents, patent sentiment analysis, patent information retrieval

Procedia PDF Downloads 84

Authors: Faith-Michael E. Uzoka, Boluwaji A. Akinnuwesi, Taiwo Amoo, Flora Aladi, Stephen Fashoto, Moses Olaniyan, Joseph Osuji

Abstract:

The overarching aim of this study is to develop a soft-computing system for the differential diagnosis of tropical diseases. These conditions are of concern to health bodies, physicians, and the community at large because of their mortality rates, and difficulties in early diagnosis due to the fact that they present with symptoms that overlap, and thus become ‘confusable’. We report on the first phase of our study, which focuses on the development of a fuzzy cognitive map model for early differential diagnosis of tropical diseases. We used malaria as a case disease to show the effectiveness of the FCM technology as an aid to the medical practitioner in the diagnosis of tropical diseases. Our model takes cognizance of manifested symptoms and other non-clinical factors that could contribute to symptoms manifestations. Our model showed 85% accuracy in diagnosis, as against the physicians’ initial hypothesis, which stood at 55% accuracy. It is expected that the next stage of our study will provide a multi-disease, multi-symptom model that also improves efficiency by utilizing a decision support filter that works on an algorithm, which mimics the physician’s diagnosis process.

Keywords: medical diagnosis, tropical diseases, fuzzy cognitive map, decision support filters, malaria differential diagnosis

Procedia PDF Downloads 313

Authors: Laura Cami Vorpsi

Abstract:

This research paper examines the potential of private law as a means to promote and safeguard human rights while also addressing the associated challenges and limitations of adopting such an approach. Historically, private law mechanisms, namely contract law, tort law, and property law, have been employed to govern and oversee private relationships and transactions. Nevertheless, it is increasingly acknowledged that private law can also assume a significant role in safeguarding and advancing human rights, particularly in circumstances where the safeguards provided by public law are insufficient or inaccessible. This study assesses the benefits associated with the utilization of private law as a complementary measure to public law safeguards. These advantages encompass enhanced efficacy and efficiency of remedies, as well as the capacity to customize solutions to suit the unique requirements and circumstances of individuals. Nevertheless, the present study also considers the constraints associated with private law mechanisms, such as the financial and procedural intricacies of legal proceedings, the possibility of imbalanced negotiation power, and the potential to worsen pre-existing disparities and systemic inequities. The paper posits that the adoption of a private law-based approach to human rights necessitates a meticulous design and implementation process in order to mitigate potential risks and optimize the advantages. In conclusion, this study examines the ramifications of these discoveries on policy and practice, highlighting the necessity for heightened awareness and education regarding the capacity of private law to advance and safeguard human rights. Additionally, it underscores the significance of establishing efficient and easily accessible mechanisms for upholding human rights within the private domain. The paper concludes by providing recommendations for future research in this domain, specifically emphasizing the necessity for additional empirical investigations to assess the efficacy and consequences of private law-oriented strategies in safeguarding human rights.

Keywords: private law, human rights, promoting, protecting, access to justice

Procedia PDF Downloads 68

Authors: Amirhossein Chambari

Abstract:

This article proposes a bi-objective model for the facility location problem subject to congestion (overcrowding). Motivated by implementations to locate servers in internet mirror sites, communication networks, one-server-systems, so on. This model consider for situations in which immobile (or fixed) service facilities are congested (or queued) by stochastic demand to behave as M/M/1/K queues. We consider for this problem two simultaneous perspectives; (1) Customers (desire to limit times of accessing and waiting for service) and (2) Service provider (desire to limit average facility idle-time). A bi-objective model is setup for facility location problem with two objective functions; (1) Minimizing sum of expected total traveling and waiting time (customers) and (2) Minimizing the average facility idle-time percentage (service provider). The proposed model belongs to the class of mixed-integer nonlinear programming models and the class of NP-hard problems. In addition, to solve the model, controlled elitist non-dominated sorting genetic algorithms (Controlled NSGA-II) and controlled elitist non-dominated ranking genetic algorithms (NRGA-I) are proposed. Furthermore, the two proposed metaheuristics algorithms are evaluated by establishing standard multiobjective metrics. Finally, the results are analyzed and some conclusions are given.

Keywords: bi-objective, facility location, queueing, controlled NSGA-II, NRGA-I

Procedia PDF Downloads 575

Authors: Chellammal Surianarayanan

Abstract:

Healthcare industry is generating large amounts of data driven by various needs such as record keeping, physician’s prescription, medical imaging, sensor data, Electronic Patient Record(EPR), laboratory, pharmacy, etc. Healthcare data is so big and complex that they cannot be managed by conventional hardware and software. The complexity of healthcare big data arises from large volume of data, the velocity with which the data is accumulated and different varieties such as structured, semi-structured and unstructured nature of data. Despite the complexity of big data, if the trends and patterns that exist within the big data are uncovered and analyzed, higher quality healthcare at lower cost can be provided. Hadoop is an open source software framework for distributed processing of large data sets across clusters of commodity hardware using a simple programming model. The core components of Hadoop include Hadoop Distributed File System which offers way to store large amount of data across multiple machines and MapReduce which offers way to process large data sets with a parallel, distributed algorithm on a cluster. Hadoop ecosystem also includes various other tools such as Hive (a SQL-like query language), Pig (a higher level query language for MapReduce), Hbase(a columnar data store), etc. In this paper an analysis has been done as how healthcare big data can be processed and analyzed using Hadoop ecosystem.

Keywords: big data analytics, Hadoop, healthcare data, towards quality healthcare

Procedia PDF Downloads 408

Authors: David Shengelia, Tamara Tsutsunava, Manana Togonidze, Giorgi Chichinadze, Giorgi Beridze

Abstract:

Orthoclase gabbro intrusive exposes in the Eastern part of the Dzirula crystalline massif of the Central Transcaucasian microcontinent. It is intruded in the Baikal quartz-diorite gneisses as a stock-like body. The intrusive is characterized by heterogeneity of rock composition: variability of mineral content and irregular distribution of rock-forming minerals. The rocks are represented by pyroxenites, gabbro-pyroxenites and gabbros of different composition – K-feldspar, pyroxene-hornblende and biotite bearing varieties. Scientific views on the genesis and age of the orthoclase gabbro intrusive are considerably different. Based on the long-term pertogeochemical and geochronological investigations of the intrusive with such an extraordinary composition the authors came to the following conclusions. According to geological and geophysical data, it is stated that in the Saurian orogeny horizontal tectonic layering of the Earth’s crust of the Central Transcaucasian microcontinent took place. That is precisely this fact that explains the formation of the orthoclase gabbro intrusive. During the tectonic doubling of the Earth’s crust of the mentioned microcontinent thick tectonic nappes of mafic and sialic layers overlap the sialic basement (‘inversion’ layer). The initial magma of the intrusive was of high-temperature basite-ultrabasite composition, crystallization products of which are pyroxenites and gabbro-pyroxenites. Petrochemical data of the magma attest to its formation in the Upper mantle and partially in the ‘crustal astenolayer’. Then, a newly formed overheated dry magma with phenocrysts of clinopyrocxene and basic plagioclase intruded into the ‘inversion’ layer. From the new medium it was enriched by the volatile components causing the selective melting and as a result the formation of leucocratic quartz-feldspar material. At the same time in the basic magma intensive transformation of pyroxene to hornblende was going on. The basic magma partially mixed with the newly formed acid magma. These different magmas intruded first into the allochthonous basite layer without its significant transformation and then into the upper sialic layer and crystallized here at a depth of 7-10 km. By petrochemical data the newly formed leucocratic granite magma belongs to the S type granites, but the above mentioned mixed magma – to H (hybrid) type. During the final stage of magmatic processes the gabbroic rocks impregnated with high-temperature feldspar-bearing material forming anorthoclase or orthoclase. Thus, so called ‘orthoclase gabbro’ includes the rocks of various genetic groups: 1. protolith of gabbroic intrusive; 2. hybrid rock – K-feldspar gabbro and 3. leucocratic quartz-feldspar bearing rock. Petrochemical and geochemical data obtained from the hybrid gabbro and from the inrusive protolith differ from each other. For the identification of petrogenetic model of the orthoclase gabbro intrusive formation LA-ICP-MS- U-Pb zircon dating has been conducted in all three genetic types of gabbro. The zircon age of the protolith – mean 221.4±1.9 Ma and of hybrid K-feldspar gabbro – mean 221.9±2.2 Ma, records crystallization time of the intrusive, but the zircon age of quartz-feldspar bearing rocks – mean 323±2.9 Ma, as well as the inherited age (323±9, 329±8.3, 332±10 and 335±11 Ma) of hybrid K-feldspar gabbro corresponds to the formation age of Late Variscan granitoids widespread in the Dzirula crystalline massif.

Keywords: The Caucasus, isotope dating, orthoclase-bearing gabbro, petrogenetic model

Procedia PDF Downloads 340

Authors: Zakieh Piri, Shahla Damanabi, Peyman Rezaii Hachesoo

Abstract:

Introduction: Healthcare organizations like other organizations suffer from a number of disorders such as Business Sponsor Disorder, Business Acceptance Disorder, Cultural/Political Disorder, Data Disorder, etc. As quality in healthcare care mostly depends on the quality of data, we aimed to identify data disorders and its symptoms in two teaching hospitals. Methods: Using a self-constructed questionnaire, we asked 20 questions in related to quality and usability of patient data stored in patient records. Research population consisted of 150 managers, physicians, nurses, medical record staff who were working at the time of study. We also asked their views about the symptoms and treatments for any data disorders they mentioned in the questionnaire. Using qualitative methods we analyzed the answers. Results: After classifying the answers, we found six main data disorders: incomplete data, missed data, late data, blurred data, manipulated data, illegible data. The majority of participants believed in their important roles in treatment of data disorders while others believed in health system problems. Discussion: As clinicians have important roles in producing of data, they can easily identify symptoms and disorders of patient data. Health information managers can also play important roles in early detection of data disorders by proactively monitoring and periodic check-ups of data.

Keywords: data disorders, quality, healthcare, treatment

Procedia PDF Downloads 427

Authors: Byron Joseph A. Hallar, Annjeannette Alain D. Galang, Maria Visitacion N. Gumabay

Abstract:

One of the unique challenges provided by the twenty-first century to Philippine higher education is the utilization of Big Data. The higher education system in the Philippines is generating burgeoning amounts of data that contains relevant data that can be used to generate the information and knowledge needed for accurate data-driven decision making. This study examines the status, relevance and future of Big Data and Analytics in Philippine higher education. The insights gained from the study may be relevant to other developing nations similarly situated as the Philippines.

Keywords: big data, data analytics, higher education, republic of the philippines, assessment

Procedia PDF Downloads 341

Authors: F. Lazzeri, I. Reiter

Abstract:

Energy production optimization has been traditionally very important for utilities in order to improve resource consumption. However, load forecasting is a challenging task, as there are a large number of relevant variables that must be considered, and several strategies have been used to deal with this complex problem. This is especially true also in microgrids where many elements have to adjust their performance depending on the future generation and consumption conditions. The goal of this paper is to present a solution for short-term load forecasting in microgrids, based on three machine learning experiments developed in R and web services built and deployed with different components of Cortana Intelligence Suite: Azure Machine Learning, a fully managed cloud service that enables to easily build, deploy, and share predictive analytics solutions; SQL database, a Microsoft database service for app developers; and PowerBI, a suite of business analytics tools to analyze data and share insights. Our results show that Boosted Decision Tree and Fast Forest Quantile regression methods can be very useful to predict hourly short-term consumption in microgrids; moreover, we found that for these types of forecasting models, weather data (temperature, wind, humidity and dew point) can play a crucial role in improving the accuracy of the forecasting solution. Data cleaning and feature engineering methods performed in R and different types of machine learning algorithms (Boosted Decision Tree, Fast Forest Quantile and ARIMA) will be presented, and results and performance metrics discussed.

Keywords: time-series, features engineering methods for forecasting, energy demand forecasting, Azure Machine Learning

Procedia PDF Downloads 289

Authors: G. Julius P. Roy, Prateek Saxena, Sanjeev Singh

Abstract:

Power distribution utilities two decades ago would collect data from its customers not later than a period of at least one month. The origin of SmartGrid and AMI has subsequently increased the sampling frequency leading to 1000 to 10000 fold increase in data quantity. This increase is notable and this steered to coin the tern Big Data in utilities. Power distribution industry is one of the largest to handle huge and complex data for keeping history and also to turn the data in to significance. Majority of the utilities around the globe are adopting SmartGrid technologies as a mass implementation and are primarily focusing on strategic interdependence and synergies of the big data coming from new information sources like AMI and intelligent SCADA, there is a rising need for new models of data management and resurrected focus on analytics to dissect data into descriptive, predictive and dictatorial subsets. The goal of this paper is to is to bring load disaggregation into smart energy toolkit for commercial usage.

Keywords: data management, analytics, energy data analytics, smart grid, smart utilities

Procedia PDF Downloads 774

Authors: Helena Maragou

Abstract:

Introducing Toni Morrison’s fiction to a classroom overseas entails a significant pedagogical investment, from monitoring students’ uncertain journey through Morrison’s shifty semantics to filling in the gaps of cultural knowledge and understanding for the students to be able to relate text to context. A rewarding process, as Morrison’s works present a tremendous opportunity for transnational dialogue, an opportunity that hinges upon Toni Morrison’s bringing to the fore the untold and unspeakable lives of racial ‘Others’, but also, crucially, upon her broader critique of Western ideological hegemony. This critique is a fundamental aspect of Toni Morrison’s politics and one that appeals to young readers of Toni Morrison in Greece at a time when the questioning of institutions and ideological traditions is precipitated by regional and global change. It is more or less self-evident that to help a class of international students get aboard a Morrison novel, an instructor should begin by providing them with cultural context. These days, students’ exposure to Hollywood representations of the African American past and present, as well as the use of documentaries, photography, music videos, etc., as supplementary class material, provide a starting point, a workable historical and cultural framework for textual comprehension. The true challenge, however, lies ahead: it is one thing for students to intellectually grasp the historical hardships and traumas of Morrison’s characters and to even engage in aesthetic appreciation of Morrison’s writing; quite another to relate to her works as articulations of experiences akin to their own. The great challenge, then, is in facilitating students’ discovery of the universal Morrison, the author who speaks across cultures while voicing the untold tales of her own people; this process of discovery entails, on a pedagogical level, that students be guided through the works’ historical context, to plunge into the intricacies of Morrison’s discourse, itself an elaborate linguistic booby trap, so as to be finally brought to reconsider their own historical experiences using the lens of Morrison’s fiction. The paper will be based on experience of teaching a Toni Morrison seminar to a class of Greek students at the American College of Greece and will draw from students’ exposure and responses to Toni Morrison’s “Nobel Prize Lecture,” as well as her novels Song of Solomon and Home.

Keywords: toni morrison, international classroom, pedagogy, African American literature

Procedia PDF Downloads 78

Authors: P. Srinivasa Rao, K. Venkatesh Sharma, G. Sadhya Devi, V. Nagesh

Abstract:

Privacy Preserving Data Publication is the main concern in present days because the data being published through the internet has been increasing day by day. This huge amount of data was named as Big Data by its size. This project deals the privacy preservation in the context of Big Data using a data warehousing solution called hive. We implemented Nearest Similarity Based Clustering (NSB) with Bottom-up generalization to achieve (v,l)-anonymity. (v,l)-Anonymity deals with the sensitivity vulnerabilities and ensures the individual privacy. We also calculate the sensitivity levels by simple comparison method using the index values, by classifying the different levels of sensitivity. The experiments were carried out on the hive environment to verify the efficiency of algorithms with Big Data. This framework also supports the execution of existing algorithms without any changes. The model in the paper outperforms than existing models.

Keywords: sensitivity, sensitive level, clustering, Privacy Preserving Data Publication (PPDP), bottom-up generalization, Big Data

Procedia PDF Downloads 288

Authors: Kemka Ihemelandu, Chukwuemeka Ihemelandu

Abstract:

Melanoma remains a public health crisis, with incidence rates increasing rapidly in the past decades. Improving diagnostic accuracy to decrease misdiagnosis using Artificial intelligence (AI) continues to be documented. Unfortunately, unintended racially biased outcomes, a product of lack of diversity in the dataset used, with a noted class imbalance favoring lighter vs. darker skin tone, have increasingly been recognized as a problem.Resulting in noted limitations of the accuracy of the Convolutional neural network (CNN)models. CNN models are prone to biased output due to biases in the dataset used to train them. Our aim in this study was the optimization of convolutional neural network algorithms to mitigate bias in the automated diagnosis of melanoma. We hypothesized that our proposed training algorithms based on a data augmentation method to optimize the diagnostic accuracy of a CNN classifier by generating new training samples from the original ones will reduce bias in the automated diagnosis of melanoma. We applied geometric transformation, including; rotations, translations, scale change, flipping, and shearing. Resulting in a CNN model that provided a modifiedinput data making for a model that could learn subtle racial features. Optimal selection of the momentum and batch hyperparameter increased our model accuracy. We show that our augmented model reduces bias while maintaining accuracy in the automated diagnosis of melanoma.

Keywords: bias, augmentation, melanoma, convolutional neural network

Procedia PDF Downloads 204

Authors: Narmeen Ali, Muhammad Arshad

Abstract:

Cohesion has an exact association with family functioning and enmeshment (togetherness) on one side and disengagement (separateness) on the other. Family cohesion can apprehend as a concerned association that family members have with each other and an affirmation of association inside the family. Family cohesion, assigned as the level of congruity or sympathetic or emotional attachment that relatives have toward each other, and it was seen to be associated with relational well-being and feeling of comfort in the young generation. The cross-sectional research design was used by the researcher to answer the research questions. A stratified sampling technique was used to collect the data from the participants. The data was collected equally from the males and females of different universities and different departments of Lahore, Pakistan. A self-report questionnaire was developed of given literature and which were found to be associated with family cohesion, interpersonal difficulties and mental health problems of university students. The demographic information included age, gender, university’s name, class, family system, parent’s education, parent’s profession, number of siblings and birth order. Correlation shows the negative relation between balanced cohesion and interpersonal difficulties, while interpersonal difficulties have a highly positive relationship with mental health problems. Mental health problems also have a negative correlation with the balanced family cohesion. Gender, family system, depression and anxiety are the significant predictors of interpersonal difficulties scale in university students. And gender showed a significant difference regarding family cohesion and interpersonal difficulty scale, as women reported more interpersonal difficulties than men.

Keywords: family cohesion, interpersonal difficulties, mental health problems, university students

Procedia PDF Downloads 118

Authors: Behnam Tavakkol

Abstract:

Uncertain data mining algorithms use different ways to consider uncertainty in data such as by representing a data object as a sample of points or a probability distribution. Fuzzy methods have long been used for clustering traditional (certain) data objects. They are used to produce non-crisp cluster labels. For uncertain data, however, besides some uncertain fuzzy k-medoids algorithms, not many other fuzzy clustering methods have been developed. In this work, we develop a fuzzy kernel k-medoids algorithm for clustering uncertain data objects. The developed fuzzy kernel k-medoids algorithm is superior to existing fuzzy k-medoids algorithms in clustering data sets with non-linearly separable clusters.

Keywords: clustering algorithm, fuzzy methods, kernel k-medoids, uncertain data

Procedia PDF Downloads 210

Authors: Amean A. Yasir, Zainab Kh. A. Al-Mahdi Al-Amean

Abstract:

In both developed and developing countries, obesity among women is increasing, but in different patterns and at very different speeds. It may have a negative effect on health, leading to reduced life expectancy and/or increased health problems. This research studied the age distribution among obese women, the types of overweight and obesity, and the extent of the problem of overweight/obesity and the obesity etiological factors among women in Hillah city in central Iraq. A total of 322 overweight and obese women were included in the study, those women were randomly selected. The Body Mass Index was used as indicator for overweight/ obesity. The incidence of overweight/obesity among age groups were estimated, the etiology factors included genetic, environmental, genetic/environmental and endocrine disease. The overweight and obese women were screened for incidence of infection and/or diseases. The study found that the prevalence of 322 overweight and obese women in Hillah city in central Iraq was 19.25% and 80.78%, respectively. The obese women types were recorded based on BMI and WHO classification as class-1 obesity (29.81%), class-2 obesity (24.22%) and class-3 obesity (26.70%), the result was discrepancy non-significant, P value < 0.05. The incidence of overweight in women was high among those aged 20-29 years (90.32%), 6.45% aged 30-39 years old and 3.22% among ≥ 60 years old, while the incidence of obesity was 20.38% for those in the age group 20-29 years, 17.30% were 30-39 years, 23.84% were 40-49 years, 16.92% were 50-59 years group and 21.53% were ≥ 60 years age group. These results confirm that the age can be considered as a significant factor for obesity types (P value < 0.0001). The result also showed that the both genetic factors and environmental factors were responsible for incidents of overweight or obesity (84.78%) p value < 0.0001. The results also recorded cases of different repeated infections (skin infection, recurrent UTI and influenza), cancer, gallstones, high blood pressure, type 2 diabetes, and infertility. Weight stigma and bias generally refers to negative attitudes; Obesity can affect quality of life, and the results of this study recorded depression among overweight or obese women. This can lead to sexual problems, shame and guilt, social isolation and reduced work performance. Overweight and Obesity are real problems among women of all age groups and is associated with the risk of diseases and infection and negatively affects quality of life. This result warrants further studies into the prevalence of obesity among women in Hillah City in central Iraq and the immune response of obese women.

Keywords: obesity, overweight, Iraq, body mass index

Procedia PDF Downloads 382

Authors: Nirmala Gopal, Sheetal Bhoola, Audecious Mugwagwa

Abstract:

This paper discusses the continued infringement and exploitation of data by non-state actors for destructive purposes, emphasizing radical terrorist organizations. It will discuss how terrorist organizations access and use data to foster their nefarious agendas. It further examines how cybersecurity, designed as a tool to curb data exploitation, is ineffective in raising global citizens' concerns about how their data can be kept safe and used for its acquired purpose. The study interrogates several policies and data protection instruments, such as the Data Protection Act, Cyber Security Policies, Protection of Personal Information(PPI) and General Data Protection Regulations (GDPR), to understand data use and storage in democratic states. The study outcomes point to the fact that international cybersecurity and cybercrime legislation, policies, and conventions have not curbed violations of data access and use by radical terrorist groups. The study recommends ways to enhance cybersecurity and reduce cyber risks using democratic principles.

Keywords: cybersecurity, data exploitation, terrorist organizations, data democracy

Procedia PDF Downloads 197

Authors: Eugenia Jilinguirian

Abstract:

In the healthcare industry, data mining is essential since it transforms the field by collecting useful data from large datasets. Data mining is the process of applying advanced analytical methods to large patient records and medical histories in order to identify patterns, correlations, and trends. Healthcare professionals can improve diagnosis accuracy, uncover hidden linkages, and predict disease outcomes by carefully examining these statistics. Additionally, data mining supports personalized medicine by personalizing treatment according to the unique attributes of each patient. This proactive strategy helps allocate resources more efficiently, enhances patient care, and streamlines operations. However, to effectively apply data mining, however, and ensure the use of private healthcare information, issues like data privacy and security must be carefully considered. Data mining continues to be vital for searching for more effective, efficient, and individualized healthcare solutions as technology evolves.

Keywords: data mining, healthcare, big data, individualised healthcare, healthcare solutions, database

Procedia PDF Downloads 61

Authors: Yunus Doğan, Ahmet Durap

Abstract:

Coastal regions are the one of the most commonly used places by the natural balance and the growing population. In coastal engineering, the most valuable data is wave behaviors. The amount of this data becomes very big because of observations that take place for periods of hours, days and months. In this study, some statistical methods such as the wave spectrum analysis methods and the standard statistical methods have been used. The goal of this study is the discovery profiles of the different coast areas by using these statistical methods, and thus, obtaining an instance based data set from the big data to analysis by using data mining algorithms. In the experimental studies, the six sample data sets about the wave behaviors obtained by 20 minutes of observations from Mersin Bay in Turkey and converted to an instance based form, while different clustering techniques in data mining algorithms were used to discover similar coastal places. Moreover, this study discusses that this summarization approach can be used in other branches collecting big data such as medicine.

Keywords: clustering algorithms, coastal engineering, data mining, data summarization, statistical methods

Procedia PDF Downloads 357

Authors: Takwa Jahan

Abstract:

Music can be used in ESL classroom to create a learning environment. As literature abounds with positive statements, music can be used as a vehicle for second language acquisition. Music can be applied as an instrument to help second language learners to acquire vocabulary, grammar, spelling and other four skills and to expand cultural knowledge. Vocabulary learning is perceived boring by learners. As listening to music and singing songs are enjoyable to students, it can be used effectively to acquire vocabulary in second language. This paper reports a study to find out how music exhilarates vocabulary acquisition as the learners stay relaxed and thus learning becomes more enjoyable. For conducting my research two groups of fifty students- music and non-music group were formed. Data were collected through class observation, test, questionnaires, and interview. The finding shows that music group acquired much amount of vocabulary than the non-music group. They enjoyed vocabulary learning activities based on listening songs.

Keywords: effective instrument, ESL classroom, music, relax environment, vocabulary learning

Procedia PDF Downloads 363

Authors: Sunit Kumar Sahoo, Prakash Chandra Senapati

Abstract:

The main drawback to design drug delivery systems with BCS class II drugs is their low bioavailabilty due to their inherent low permeability characteristics. So the present investigation aspire to develop a self-nanoemulsifying drug delivery system (SNEDDS) of BCS class II anti HIV drug efavirenz (EFZ) using mixtures of non-ionic surfactant mixtures with the main objective to improve the oral bioavailability of said drug. Results obtained from solubility studies of EFZ in various expients utilized for construction of the pseudo ternary phase diagram containing surfactant mixtures. Surfactants in 1:1 combination are used with different co-surfactants in different ratio to delineate the area of monophasic region of the pseudo ternary phase diagram. The formulations which offered positive results in different thermodynamic stability studies were considered for percentage transmittance and turbidity analysis. The various characterization studies like the TEM analysis of post diluted SNEDDS formulations r confirmed the size in nanometric range (below 50 nm) and FT-IR studies confirmed the intactness of the drug the in the preconcentrate. The in vitro dissolution profile of SNEDDS showed that 80% drug was released within 30 min in case of optimized SNEDDS while it was approximately 18.3 % in the case of plain drug powder.. The Pharmacokinetic study using rat model revealed a 2.63 fold increase in AUC (0-∞) in comparison to plain EFZ suspension. The designed delivery system illustrated the confidence in creating a formulation of EFZ with enhanced bioavailability for better HIV treatment.

Keywords: efavirenz, self-nanoemulsifying, surfactant mixture, bioavailability

Procedia PDF Downloads 347

Authors: Mark Anthony Casimiro, Ivan Culaba, Cornelia Soto

Abstract:

Misconceptions are the wrong concepts understood by the students which may come up based on what they experience and observe around their environment. This seemed to hinder students’ learning. In this study, six different misconceptions were determined by the researcher from the previous researches. Teachers play a vital role in the classroom. The use of appropriate strategies can contribute a lot in the success of teaching and learning Physics. The current study aimed to compare two strategies- Interactive Lecture Demonstration (ILD) and Inquiry-Based Instruction (IBI) in addressing students’ misconceptions in electric circuits. These two strategies are both interactive learning activities and student-centered. In ILD, the teacher demonstrates the activity and the students have their predictions while in IBI, students perform the experiments. The study used the mixed method in which quantitative and qualitative researches were combined. The main data of this study were the test scores of the students from the pretest and posttest. Likewise, an interview with the teacher, observer and students was done before, during and after the execution of the activities. Determining and Interpreting Resistive Electric Circuits Test version 2 (DIRECT v.2) was the instrument used in the study. Two sections of Grade 9 students from Kalumpang National High School were the respondents of the study. The two strategies were executed to each section; one class was assigned as the ILD group and the other class was the IBI group. The Physics teacher of the said school was the one who taught and executed the activities. The researcher taught the teacher the steps in doing the two strategies. The Department of Education level of proficiency in the Philippines was adopted in scoring and interpretation. The students’ level of proficiency was used in assessing students’ knowledge on electric circuits. The pretest result of the two groups had a p-value of 0.493 which was greater than the level of significance 0.05 (p >0.05) and it implied that the students’ level of understanding in the topic was the same before the execution of the strategies. The posttest results showed that the p-value (0.228) obtained was greater than the level of significance which is 0.05 (p> 0.05). This implied that the students from the ILD and IBI groups had the same level of understanding after the execution of the two strategies. This could be inferred that either of the two strategies- Interactive Lecture Demonstration and Inquiry-Based Instruction could be used in addressing students’ misconception in electric circuit as both had similar effect on the students’ level of understanding in the topic. The result of this study may greatly help teachers, administration, school heads think of appropriate strategies that can address misconceptions depending on the availability of their materials of their school.

Keywords: inquiry- based instruction, interactive lecture demonstration, misconceptions, mixed method

Procedia PDF Downloads 216

Authors: Mansour Mohieddin Ghomshei, Reza Shahi

Abstract:

Laminated composite tubes with adhesively bonded joints are widely used in aerospace and automotive industries as well as oil and gas industries. In this research, adhesively tubular single lap joints subjected to torsional and hygrothermal loadings are studied using the differential quadrature method (DQM). The analysis is based on the classical shell theory. At first, an approximate closed form solution is developed by omitting the lateral deflections in the connecting tubes. Using the analytical model, the circumferential displacements in tubes and the shear stresses in the interfacing adhesive layer are determined. Then, a numerical formulation is presented using DQM in which the lateral deflections are taken into account. By using the DQM formulation, the circumferential and radial displacements in tubes as well as shear and peel stresses in the adhesive layer are calculated. Results obtained from the proposed DQM solutions are compared well with those of the approximate analytical model and those of some published references. Finally using the DQM model, parametric studies are carried out to investigate the influence of various parameters such as adhesive layer thickness, torsional loading, overlap length, tubes radii, relative humidity, and temperature.

Keywords: adhesively bonded joint, differential quadrature method (DQM), hygrothermal, laminated composite tube

Procedia PDF Downloads 296

Authors: Anny Retnowati, Elisabeth Sundari

Abstract:

This research aims to elaborate the meaning of personal data protection principles on patient access to health data in medical records in Indonesia and its implications. The method uses normative legal research by examining health law in Indonesia regarding the patient's right to access their health data in medical records. The data will be analysed qualitatively using the interpretation method to elaborate on the limitation of the meaning of personal data protection principles on patients' access to their data in medical records. The results show that patients only have the right to obtain copies of their health data in medical records. There is no right to inspect directly at any time. Indonesian health law limits the principle of patients' right to broad access to their health data in medical records. This restriction has implications for the reduction of personal data protection as part of human rights. This research contribute to show that a limitaion of personal data protection may abuse the human rights.

Keywords: access, health data, medical records, personal data, protection

Procedia PDF Downloads 82