Search results for: forecasting accuracy

Authors: Mingyang Sun, Guozheng Yan, Dasheng Liu, Lei Yang

Abstract:

Driven by the demand of intelligent monitoring in rehabilitation centers or hospitals, a high accuracy real-time location system based on UWB (ultra-wideband) technology was proposed. The system measures precise location of a specific person, traces his movement and visualizes his trajectory on the screen for doctors or administrators. Therefore, doctors could view the position of the patient at any time and find them immediately and exactly when something emergent happens. In our design process, different algorithms were discussed, and their errors were analyzed. In addition, we discussed about a , simple but effective way of correcting the antenna delay error, which turned out to be effective. By choosing the best algorithm and correcting errors with corresponding methods, the system attained a good accuracy. Experiments indicated that the ranging error of the system is lower than 7 cm, the locating error is lower than 20 cm, and the refresh rate exceeds 5 times per second. In future works, by embedding the system in wearable IoT (Internet of Things) devices, it could provide not only physical parameters, but also the activity status of the patient, which would help doctors a lot in performing healthcare.

Keywords: intelligent monitoring, ultra-wideband technology, real-time location, IoT devices, smart healthcare

Procedia PDF Downloads 127

Authors: Hamed Hossein Afshari, Shahrzad Jalali, Amir Hossein Ghods, Bijan Raahemi

Abstract:

This paper introduces an automated clustering solution that applies to WiFi/Bluetooth sensing data and is later used for traffic management applications. The paper initially summarizes a number of clustering approaches and thereafter shows their performance for noise removal. In this context, clustering is used to recognize WiFi and Bluetooth MAC addresses that belong to passengers traveling by a public urban transit bus. The main objective is to build an intelligent system that automatically filters out MAC addresses that belong to persons located outside the bus for different routes in the city of Ottawa. The proposed intelligent system alleviates the need for defining restrictive thresholds that however reduces the accuracy as well as the range of applicability of the solution for different routes. This paper moreover discusses the performance benefits of the presented clustering approaches in terms of the accuracy, time and space complexity, and the ease of use. Note that results of clustering can further be used for the purpose of the origin-destination estimation of individual passengers, predicting the traffic load, and intelligent management of urban bus schedules.

Keywords: WiFi-Bluetooth sensing, cluster analysis, artificial intelligence, traffic management

Procedia PDF Downloads 225

Authors: Alex Eldo Simon, Abhishek Yadav

Abstract:

This study aimed to evaluate the utility of postmortem computed tomography (CT) and heart weight measurements in the assessment of cardiomegaly in cases of sudden death due to cardiac origin by comparing the results of these two diagnostic methods. The study retrospectively analyzed postmortem computed tomography (PMCT) data from 54 cases of sudden natural death and compared the findings with those of the autopsy. The study involved measuring the cardiothoracic ratio (CTR) from coronal computed tomography (CT) images and determining the actual cardiac weight by weighing the heart during the autopsy. The inclusion criteria for the study were cases of sudden death suspected to be caused by cardiac pathology, while exclusion criteria included death due to unnatural causes such as trauma or poisoning, diagnosed natural causes of death related to organs other than the heart, and cases of decomposition. Sensitivity, specificity, and diagnostic accuracy were calculated, and to evaluate the accuracy of using the cardiothoracic ratio (CTR) to detect an enlarged heart, the study generated receiver operating characteristic (ROC) curves. The cardiothoracic ratio (CTR) is a radiological tool used to assess cardiomegaly by measuring the maximum cardiac diameter in relation to the maximum transverse diameter of the chest wall. The clinically used criteria for CTR have been modified from 0.50 to 0.57 for use in postmortem settings, where abnormalities can be detected by comparing CTR values to this threshold. A CTR value of 0.57 or higher is suggestive of hypertrophy but not conclusive. Similarly, heart weight is measured during the traditional autopsy, and a cardiac weight greater than 450 grams is defined as hypertrophy. Of the 54 cases evaluated, 22 (40.7%) had a cardiothoracic ratio (CTR) ranging from > 0.50 to equal 0.57, and 12 cases (22.2%) had a CTR greater than 0.57, which was defined as hypertrophy. The mean CTR was calculated as 0.52 ± 0.06. Among the 54 cases evaluated, the weight of the heart was measured, and the mean was calculated as 369.4 ± 99.9 grams. Out of the 54 cases evaluated, 12 were found to have hypertrophy as defined by PMCT, while only 9 cases were identified with hypertrophy in traditional autopsy. The sensitivity and specificity of the test were calculated as 55.56% and 84.44%, respectively. The sensitivity of the hypertrophy test was found to be 55.56% (95% CI: 26.66, 81.12¹), the specificity was 84.44% (95% CI: 71.22, 92.25¹), and the diagnostic accuracy was 79.63% (95% CI: 67.1, 88.23¹). The limitation of the study was a low sample size of only 54 cases, which may limit the generalizability of the findings. The comparison of the cardiothoracic ratio with heart weight in this study suggests that PMCT may serve as a screening tool for medico-legal autopsies when performed by forensic pathologists. However, it should be noted that the low sensitivity of the test (55.5%) may limit its diagnostic accuracy, and therefore, further studies with larger sample sizes and more diverse populations are needed to validate these findings.

Keywords: PMCT, virtopsy, CTR, cardiothoracic ratio

Procedia PDF Downloads 70

Authors: Vidar Hepsø

Abstract:

The energy and oil companies on the Norwegian Continental shelf come from a situation where each asset control and manage their energy supply (island mode) and move towards a situation where the assets need to collaborate and coordinate energy use with others due to increased cost and scarcity of electric energy sharing the energy that is provided. Currently, several areas are electrified either with an onshore grid cable or are receiving intermittent energy from offshore wind-parks. While the onshore grid in Norway is well regulated, the offshore grid is still in the making, with several oil and gas electrification projects and offshore wind development just started. The paper will describe the shift in the mindset that comes with operating this new offshore grid. This transition process heralds an increase in collaboration across boundaries and integration of energy management across companies, businesses, technical disciplines, and engagement with stakeholders in the larger society. This transition will be described as a function of the new challenges with increased complexity of the energy mix (wind, oil/gas, hydrogen and others) coupled with increased technical and organization complexity in energy management. Organizational complexity denotes an increasing integration across boundaries, whether these boundaries are company, vendors, professional disciplines, regulatory regimes/bodies, businesses, and across numerous societal stakeholders. New practices must be developed, made legitimate and institutionalized across these boundaries. Only parts of this complexity can be mitigated technically, e.g.: by use of batteries, mixing energy systems and simulation/ forecasting tools. Many challenges must be mitigated with legitimated societal and institutionalized governance practices on many levels. Offshore electrification supports Norway’s 2030 climate targets but is also controversial since it is exploiting the larger society’s energy resources. This means that new systems and practices must also be transparent, not only for the industry and the authorities, but must also be acceptable and just for the larger society. The paper report from ongoing work in Norway, participant observation and interviews in projects and people working with offshore grid development in Norway. One case presented is the development of an offshore floating windfarm connected to two offshore installations and the second case is an offshore grid development initiative providing six installations electric energy via an onshore cable. The development of the offshore grid is analyzed using a capability platform framework, that describes the technical, competence, work process and governance capabilities that are under development in Norway. A capability platform is a ‘stack’ with the following layers: intelligent infrastructure, information and collaboration, knowledge sharing & analytics and finally business operations. The need for better collaboration and energy forecasting tools/capabilities in this stack will be given a special attention in the two use cases that are presented.

Keywords: capability platform, electrification, carbon footprint, control rooms, energy forecsting, operational model

Procedia PDF Downloads 55

Authors: Piotr Fiszeder, Marcin Fałdziński, Peter Molnár

Abstract:

The dynamic conditional correlation model of Engle (2002) is one of the most popular multivariate volatility models. However, this model is based solely on closing prices. It has been documented in the literature that the high and low price of the day can be used in an efficient volatility estimation. We, therefore, suggest a model which incorporates high and low prices into the dynamic conditional correlation framework. Empirical evaluation of this model is conducted on three datasets: currencies, stocks, and commodity exchange-traded funds. The utilisation of realized variances and covariances as proxies for true variances and covariances allows us to reach a strong conclusion that our model outperforms not only the standard dynamic conditional correlation model but also a competing range-based dynamic conditional correlation model.

Keywords: volatility, DCC model, high and low prices, range-based models, covariance forecasting

Procedia PDF Downloads 168

Authors: Jiahe Liu, Hongyang Yu, Feng Qian, Miao Luo

Abstract:

SMPL-X was a powerful parametric human body model that included male, neutral, and female models, with significant gender differences between these three models. During the process of 3D human body reconstruction, the correct selection of standard templates was crucial for obtaining accurate results. To address this issue, we developed an efficient gender classification algorithm to automatically select the appropriate template for 3D human body reconstruction. The key to this gender classification algorithm was the precise analysis of human body features. By using the SMPL-X model, the algorithm could detect and identify gender features of the human body, thereby determining which standard template should be used. The accuracy of this algorithm made the 3D reconstruction process more accurate and reliable, as it could adjust model parameters based on individual gender differences. SMPL-X and the related gender classification algorithm have brought important advancements to the field of 3D human body reconstruction. By accurately selecting standard templates, they have improved the accuracy of reconstruction and have broad potential in various application fields. These technologies continue to drive the development of the 3D reconstruction field, providing us with more realistic and accurate human body models.

Keywords: gender classification, joint detection, SMPL-X, 3D reconstruction

Procedia PDF Downloads 56

Authors: Su Yin Chiam, Qing Xin Zhang, Jaehoon Chung

Abstract:

Complementary metal-oxide-semiconductor (CMOS) integrated circuits (ICs) have obtained increased attention in the biosensor community because CMOS technology provides cost-effective and high-performance signal processing at a mass-production level. In order to supply biological samples and reagents effectively to the sensing elements, there are increasing demands for seamless integration of microfluidic components on the fabricated CMOS wafers by post-processing. Although the PDMS microfluidic channels replicated from separately prepared silicon mold can be typically aligned and bonded onto the CMOS wafers, it remains challenging owing the inherently limited aligning accuracy ( > ± 10 μm) between the two layers. Here we present a new post-processing method to create three-dimensional microfluidic components using two different polarities of photoresists, an epoxy-based negative SU-8 photoresist and positive SPR220-7 photoresist. The positive photoresist serves as a sacrificial layer and the negative photoresist was utilized as a structural material to generate three-dimensional structures. Because both photoresists are patterned using a standard photolithography technology, the dimensions of the structures can be effectively controlled as well as the alignment accuracy, moreover, is dramatically improved (< ± 2 μm) and appropriately can be adopted as an alternative post-processing method. To validate the proposed processing method, we applied this technique to build cell-trapping structures. The SU8 photoresist was mainly used to generate structures and the SPR photoresist was used as a sacrificial layer to generate sub-channel in the SU8, allowing fluid to pass through. The sub-channel generated by etching the sacrificial layer works as a cell-capturing site. The well-controlled dimensions enabled single-cell capturing on each site and high-accuracy alignment made cells trapped exactly on the sensing units of CMOS biosensors.

Keywords: SU-8, microfluidic, MEMS, microfabrication

Procedia PDF Downloads 499

Authors: Julio Manuel de Luis Ruiz, Javier Sedano Cibrián, Rubén Pérez Álvarez, Raúl Pereda García, Cristina Diego Soroa

Abstract:

Nowadays, the use of drones has been extended to practically any human activity. One of the main applications is focused on the surveying field. In this regard, software programs that process the images captured by the sensor from the drone in an almost automatic way have been developed and commercialized, but they only allow contrasting the results through control points. This work proposes the contrast of a 3D model obtained from a flight developed by a drone and a non-metric camera (due to its low cost), with a second model that is obtained by means of the historically-endorsed classical methods. In addition to this, the contrast is developed over a certain territory with a significant unevenness, so as to test the model generated with photogrammetry, and considering that photogrammetry with drones finds more difficulties in terms of accuracy in this kind of situations. Distances, heights, surfaces and volumes are measured on the basis of the 3D models generated, and the results are contrasted. The differences are about 0.2% for the measurement of distances and heights, 0.3% for surfaces and 0.6% when measuring volumes. Although they are not important, they do not meet the order of magnitude that is presented by salespeople.

Keywords: accuracy, classical topographic, model tridimensional, photogrammetry, Uav.

Procedia PDF Downloads 123

Authors: Reza Mohammadi, Mahmod R. Sahebi, Mehrnoosh Omati, Milad Vahidi

Abstract:

Classification of high resolution polarimetric Synthetic Aperture Radar (PolSAR) images plays an important role in land cover and land use management. Recently, classification algorithms based on Bag of Visual Words (BOVW) model have attracted significant interest among scholars and researchers in and out of the field of remote sensing. In this paper, BOVW model with pixel based low-level features has been implemented to classify a subset of San Francisco bay PolSAR image, acquired by RADARSAR 2 in C-band. We have used segment-based decision-making strategy and compared the result with the result of traditional Support Vector Machine (SVM) classifier. 90.95% overall accuracy of the classification with the proposed algorithm has shown that the proposed algorithm is comparable with the state-of-the-art methods. In addition to increase in the classification accuracy, the proposed method has decreased undesirable speckle effect of SAR images.

Keywords: Bag of Visual Words (BOVW), classification, feature extraction, land cover management, Polarimetric Synthetic Aperture Radar (PolSAR)

Procedia PDF Downloads 192

Authors: Eunsu Jang, Kang Park

Abstract:

In developing an armored ground combat vehicle (AGCV), it is a very important step to analyze the vulnerability (or the survivability) of the AGCV against enemy’s attack. In the vulnerability analysis, the penetration equations are usually used to get the penetration depth and check whether a bullet can penetrate the armor of the AGCV, which causes the damage of internal components or crews. The penetration equations are derived from penetration experiments which require long time and great efforts. However, they usually hold only for the specific material of the target and the specific type of the bullet used in experiments. Thus, penetration simulation using ANSYS can be another option to calculate penetration depth. However, it is very important to model the targets and select the input parameters in order to get an accurate penetration depth. This paper performed a sensitivity analysis of input parameters of ANSYS on the accuracy of the calculated penetration depth. Two conflicting objectives need to be achieved in adopting ANSYS in penetration analysis: maximizing the accuracy of calculation and minimizing the calculation time. To maximize the calculation accuracy, the sensitivity analysis of the input parameters for ANSYS was performed and calculated the RMS error with the experimental data. The input parameters include mesh size, boundary condition, material properties, target diameter are tested and selected to minimize the error between the calculated result from simulation and the experiment data from the papers on the penetration equation. To minimize the calculation time, the parameter values obtained from accuracy analysis are adjusted to get optimized overall performance. As result of analysis, the followings were found: 1) As the mesh size gradually decreases from 0.9 mm to 0.5 mm, both the penetration depth and calculation time increase. 2) As diameters of the target decrease from 250mm to 60 mm, both the penetration depth and calculation time decrease. 3) As the yield stress which is one of the material property of the target decreases, the penetration depth increases. 4) The boundary condition with the fixed side surface of the target gives more penetration depth than that with the fixed side and rear surfaces. By using above finding, the input parameters can be tuned to minimize the error between simulation and experiments. By using simulation tool, ANSYS, with delicately tuned input parameters, penetration analysis can be done on computer without actual experiments. The data of penetration experiments are usually hard to get because of security reasons and only published papers provide them in the limited target material. The next step of this research is to generalize this approach to anticipate the penetration depth by interpolating the known penetration experiments. This result may not be accurate enough to be used to replace the penetration experiments, but those simulations can be used in the early stage of the design process of AGCV in modelling and simulation stage.

Keywords: ANSYS, input parameters, penetration depth, sensitivity analysis

Procedia PDF Downloads 379

Authors: Abraham Finkelman

Abstract:

Computerized Surgical Guides have been proven to be a predictable way to perform dental implants, with a relatively high accuracy in comparison to a treatment plan. When using the CSG Bone supported, it allows us to make the necessary changes of the hard tissue prior to the implant placement and after the implant placement. The CSG gives us an accurate position for the drilling, and during the implant placement it allows us to alter the vertical position of the implant altering the final position of the abutment and avoiding any risk of any damage to the adjacent anatomical structures. Any Changes required to the bone level can be done prior to the fixation of the CSG using a reduction guide, which incur extra surgical fees and the need of a second surgical guide. Any changes of the bone level after the implant placement are at the risk of damaging the implant neck surface. The technique consists of a universal system that allows us to remove the excess bone around the implant sockets prior to the implant placement which then enables us to place the implant in the vertical position with accuracy as planned with the CSG. The systems consist of a hollow pin of different sizes and diameters. Depending on the implant system that we are using. Length sizes are from 6mm-16mm and a diameter of 2.6mm-4.8mm. Upon the completion of the drilling, the pin is then inserted into the implant socket-using the insertion tool. Once the insertion tool has unscrewed the pin, we can continue with the bone reduction. The bone reduction can be done using conventional methods upon the removal of all the excess bone around the pin. The insertion tool is then screwed into the pin and the pin is then removed. We now, have the new bone level at the crest of the implant socket which is our mark for the vertical position of the implant. In some cases, when we are locating the implant very close to anatomical structures, any form of deviation to the vertical position of the implant during the surgery, can cause damage to such anatomical structures, creating irreversible damages such as paresthesia or dysesthesia of the mandibular nerve. If we are planning for immediate loading and we have done our temporary restauration in base of our computerized plan, deviation in the vertical position of the implant will affect the position of the abutment, affecting the accuracy of the temporary prosthesis, extending the working time till we adapt the prosthesis to the new position.

Keywords: bone reduction, computer aided navigation, dental implant placement, surgical guides

Procedia PDF Downloads 319

Authors: Melaku Tsehay

Abstract:

The Consortium of Christian Relief and Development Associations (CCRDA), and the CORE Group Polio Partners (CGPP) Secretariat have been working with Global Alliance for Vac-cines and Immunization (GAVI) to improve the immunization data quality in Afar and Somali Regional States. The main aim of this study was to compare the quality of immunization data before and after the above interventions in health facilities in the pastoralist communities in Ethiopia. To this end, a comparative-cross-sectional study was conducted on 51 health facilities. The baseline data was collected in May 2019, while the end line data in August 2021. The WHO data quality self-assessment tool (DQS) was used to collect data. A significant improvment was seen in the accuracy of the pentavalent vaccine (PT)1 (p = 0.012) data at the health posts (HP), while PT3 (p = 0.010), and Measles (p = 0.020) at the health centers (HC). Besides, a highly sig-nificant improvment was observed in the accuracy of tetanus toxoid (TT)2 data at HP (p < 0.001). The level of over- or under-reporting was found to be < 8%, at the HP, and < 10% at the HC for PT3. The data completeness was also increased from 72.09% to 88.89% at the HC. Nearly 74% of the health facilities timely reported their respective immunization data, which is much better than the baseline (7.1%) (p < 0.001). These findings may provide some hints for the policies and pro-grams targetting on improving immunization data qaulity in the pastoralist communities.

Keywords: data quality, immunization, verification factor, pastoralist region

Procedia PDF Downloads 88

Authors: Tatsuya Otoshi, Masayuki Murata

Abstract:

With the development and diversification of applications on the Internet, applications that require high responsiveness, such as video streaming, are becoming mainstream. Application responsiveness is not only a matter of communication delay but also a matter of time required to grasp changes in network conditions. The tradeoff between accuracy and measurement time is a challenge in network control. We people make countless decisions all the time, and our decisions seem to resolve tradeoffs between time and accuracy. When making decisions, people are known to make appropriate choices based on relatively small samples. Although there have been various studies on models of human decision-making, a model that integrates various cognitive biases, called ”quantum decision-making,” has recently attracted much attention. However, the modeling of small samples has not been examined much so far. In this paper, we extend the model of quantum decision-making to model decision-making with a small sample. In the proposed model, the state is updated by value-based probability amplitude amplification. By analytically obtaining a lower bound on the number of samples required for decision-making, we show that decision-making with a small number of samples is feasible.

Keywords: quantum decision making, small sample, MPEG-DASH, Grover's algorithm

Procedia PDF Downloads 63

Authors: Jaishree Ranganathan, MuthuPriya Shanmugakani Velsamy, Shamika Kulkarni, Angelina Tzacheva

Abstract:

Emotions play an important role in everyday life. An-alyzing these emotions or feelings from social media platforms like Twitter, Facebook, blogs, and forums based on user comments and reviews plays an important role in various factors. Some of them include brand monitoring, marketing strategies, reputation, and competitor analysis. The opinions or sentiments mined from such data helps understand the current state of the user. It does not directly provide intuitive insights on what actions to be taken to benefit the end user or business. Actionable Pattern Mining method provides suggestions or actionable recommendations on what changes or actions need to be taken in order to benefit the end user. In this paper, we propose automatic classification of emotions in Twitter data using Recurrent Neural Network - Gated Recurrent Unit. We achieve training accuracy of 87.58% and validation accuracy of 86.16%. Also, we extract action rules with respect to the user emotion that helps to provide actionable suggestion.

Keywords: emotion mining, twitter, recurrent neural network, gated recurrent unit, actionable pattern mining

Procedia PDF Downloads 152

Authors: Marthadale Acibo Semacio

Abstract:

Speaking skill is a crucial part of English language teaching and learning. This actually shows the great importance of this skill in English language classes. Through speaking, ideas and thoughts are shared with other people, and a smooth interaction between people takes place. The study examined the levels of speaking proficiency of the control and experimental groups on pronunciation, grammatical accuracy, and fluency. As a quasi-experimental study, it also determined the presence or absence of significant changes in their speaking proficiency levels in terms of pronouncing the words correctly, the accuracy of grammar and fluency of a language given the two methods to the groups of students in the English language, using the traditional and audio-lingual methods. Descriptive and inferential statistics were employed according to the stated specific problems. The study employed a video presentation with prior information about it. In the video, the teacher acts as model one, giving instructions on what is going to be done, and then the students will perform the activity. The students were paired purposively based on their learning capabilities. Observing proper ethics, their performance was audio recorded to help the researcher assess the learner using the modified speaking rubric. The study revealed that those under the traditional method were more fluent than those in the audio-lingual method. With respect to the way in which each method deals with the feelings of the student, the audio-lingual one fails to provide a principle that would relate to this area and follows the assumption that the intrinsic motivation of the students to learn the target language will spring from their interest in the structure of the language. However, the speaking proficiency levels of the students were remarkably reinforced in reading different words through the aid of aural media with their teachers. The study concluded that using an audio-lingual method of teaching is not a stand-alone method but only an aid of the teacher in helping the students improve their speaking proficiency in the English Language. Hence, audio-lingual approach is encouraged to be used in teaching English language, on top of the chalk-talk or traditional method, to improve the speaking proficiency of students.

Keywords: audio-lingual, speaking, grammar, pronunciation, accuracy, fluency, proficiency

Procedia PDF Downloads 54

Authors: Issa Qabaja, Fadi Thabtah

Abstract:

Email phishing classification is one of the vital problems in the online security research domain that have attracted several scholars due to its impact on the users payments performed daily online. One aspect to reach a good performance by the detection algorithms in the email phishing problem is to identify the minimal set of features that significantly have an impact on raising the phishing detection rate. This paper investigate three known feature selection methods named Information Gain (IG), Chi-square and Correlation Features Set (CFS) on the email phishing problem to separate high influential features from low influential ones in phishing detection. We measure the degree of influentially by applying four data mining algorithms on a large set of features. We compare the accuracy of these algorithms on the complete features set before feature selection has been applied and after feature selection has been applied. After conducting experiments, the results show 12 common significant features have been chosen among the considered features by the feature selection methods. Further, the average detection accuracy derived by the data mining algorithms on the reduced 12-features set was very slight affected when compared with the one derived from the 47-features set.

Keywords: data mining, email classification, phishing, online security

Procedia PDF Downloads 417

Authors: Xu Han, Shanxiong Chen, Shiyu Zhu, Xiaoyu Lin, Fujia Zhao, Dingwang Wang

Abstract:

Character detection is an important issue for character recognition of ancient Yi books. The accuracy of detection directly affects the recognition effect of ancient Yi books. Considering the complex layout, the lack of standard typesetting and the mixed arrangement between images and texts, we propose a character detection method for ancient Yi books based on connected components and regressive character segmentation. First, the scanned images of ancient Yi books are preprocessed with nonlocal mean filtering, and then a modified local adaptive threshold binarization algorithm is used to obtain the binary images to segment the foreground and background for the images. Second, the non-text areas are removed by the method based on connected components. Finally, the single character in the ancient Yi books is segmented by our method. The experimental results show that the method can effectively separate the text areas and non-text areas for ancient Yi books and achieve higher accuracy and recall rate in the experiment of character detection, and effectively solve the problem of character detection and segmentation in character recognition of ancient books.

Keywords: CCS concepts, computing methodologies, interest point, salient region detections, image segmentation

Procedia PDF Downloads 116

Authors: Seyoung Kim, Jeongmin Kim, Kwang Ryel Ryu

Abstract:

A database that records average traffic speeds measured at five-minute intervals for all the links in the traffic network of a metropolitan city. While learning from this data the models that can predict future traffic speed would be beneficial for the applications such as the car navigation system, building predictive models for every link becomes a nontrivial job if the number of links in a given network is huge. An advantage of adopting k-nearest neighbor (k-NN) as predictive models is that it does not require any explicit model building. Instead, k-NN takes a long time to make a prediction because it needs to search for the k-nearest neighbors in the database at prediction time. In this paper, we investigate how much we can speed up k-NN in making traffic speed predictions by reducing the amount of data to be searched for without a significant sacrifice of prediction accuracy. The rationale behind this is that we had a better look at only the recent data because the traffic patterns not only repeat daily or weekly but also change over time. In our experiments, we build several different k-NN models employing different sets of features which are the current and past traffic speeds of the target link and the neighbor links in its up/down-stream. The performances of these models are compared by measuring the average prediction accuracy and the average time taken to make a prediction using various amounts of data.

Keywords: big data, k-NN, machine learning, traffic speed prediction

Procedia PDF Downloads 344

Authors: Maryam Heidari, James H. Jones Jr.

Abstract:

Millions of online posts about different topics and products are shared on popular social media platforms. One use of this content is to provide crowd-sourced information about a specific topic, event or product. However, this use raises an important question: what percentage of information available through these services is trustworthy? In particular, might some of this information be generated by a machine, i.e., a bot, instead of a human? Bots can be, and often are, purposely designed to generate enough volume to skew an apparent trend or position on a topic, yet the consumer of such content cannot easily distinguish a bot post from a human post. In this paper, we introduce a model for social media bot detection which uses Bidirectional Encoder Representations from Transformers (Google Bert) for sentiment classification of tweets to identify topic-independent features. Our use of a Natural Language Processing approach to derive topic-independent features for our new bot detection model distinguishes this work from previous bot detection models. We achieve 94\% accuracy classifying the contents of data as generated by a bot or a human, where the most accurate prior work achieved accuracy of 92\%.

Keywords: bot detection, natural language processing, neural network, social media

Procedia PDF Downloads 104

Authors: Megha Gupta, Nupur Prakash

Abstract:

Identification of plant diseases has been performed using machine learning and deep learning models on the datasets containing images of healthy and diseased plant leaves. The current study carries out an evaluation of some of the deep learning models based on convolutional neural network (CNN) architectures for identification of plant diseases. For this purpose, the publicly available New Plant Diseases Dataset, an augmented version of PlantVillage dataset, available on Kaggle platform, containing 87,900 images has been used. The dataset contained images of 26 diseases of 14 different plants and images of 12 healthy plants. The CNN models selected for the study presented in this paper are AlexNet, ZFNet, VGGNet (four models), GoogLeNet, and ResNet (three models). The selected models are trained using PyTorch, an open-source machine learning library, on Google Colaboratory. A comparative study has been carried out to analyze the high degree of accuracy achieved using these models. The highest test accuracy and F1-score of 99.59% and 0.996, respectively, were achieved by using GoogLeNet with Mini-batch momentum based gradient descent learning algorithm.

Keywords: comparative analysis, convolutional neural networks, deep learning, plant disease identification

Procedia PDF Downloads 183

Authors: Sabiu Bala Muhammad, Rosli Saad

Abstract:

Multiple linear regression (MLR) models for fast estimation of true subsurface resistivity from apparent resistivity field measurements are developed and assessed in this study. The parameters investigated were apparent resistivity (ρₐ), horizontal location (X) and depth (Z) of measurement as the independent variables; and true resistivity (ρₜ) as the dependent variable. To achieve linearity in both resistivity variables, datasets were first transformed into logarithmic domain following diagnostic checks of normality of the dependent variable and heteroscedasticity to ensure accurate models. Four MLR models were developed based on hierarchical combination of the independent variables. The generated MLR coefficients were applied to another data set to estimate ρₜ values for validation. Contours of the estimated ρₜ values were plotted and compared to the observed data plots at the colour scale and blanking for visual assessment. The accuracy of the models was assessed using coefficient of determination (R²), standard error (SE) and weighted mean absolute percentage error (wMAPE). It is concluded that the MLR models can estimate ρₜ for with high level of accuracy.

Keywords: apparent resistivity, depth, horizontal location, multiple linear regression, true resistivity

Procedia PDF Downloads 260

Authors: Usiobaifo Agharese Rosemary, Osaseri Roseline Oghogho

Abstract:

Diabetes is a complex group of disease with a variety of causes; it is a disorder of the body metabolism in the digestion of carbohydrates food. The application of machine learning in the field of medical diagnosis has been the focus of many researchers and the use of recognition and classification model as a decision support tools has help the medical expert in diagnosis of diseases. Considering the large volume of medical data which require special techniques, experience, and high diagnostic skill in the diagnosis of diseases, the application of an artificial intelligent system to assist medical personnel in order to enhance their efficiency and accuracy in diagnosis will be an invaluable tool. In this study will propose a diabetes diagnosis model using rough set and K-nearest Neighbor classifier algorithm. The system consists of two modules: the feature extraction module and predictor module, rough data set is used to preprocess the attributes while K-nearest neighbor classifier is used to classify the given data. The dataset used for this model was taken for University of Benin Teaching Hospital (UBTH) database. Half of the data was used in the training while the other half was used in testing the system. The proposed model was able to achieve over 80% accuracy.

Keywords: classifier algorithm, diabetes, diagnostic model, machine learning

Procedia PDF Downloads 323

Authors: Rafał Urbaniak, Robert Kłosowiak, Michał Ciałkowski, Jarosław Bartoszewicz

Abstract:

The Chair of Thermal Engineering at Poznan University of Technology has been conducted research works on the possibilities of using carbon nanostructures in energy and mechanics applications for a couple of years. Those studies have provided results in a form of co-operation with foreign research centres, numerous publications and patent applications. Authors of this paper have studied the influence of multi-walled carbon nanostructures on changes in static friction arising when steel surfaces were moved. Tests were made using the original test stand consisting of automatically controlled inclined plane driven by precise stepper motors. Computer program created in the LabView environment was responsible for monitoring of the stand operation, accuracy of measurements and archiving the obtained results. Such a solution enabled to obtain high accuracy and repeatability of all conducted experiments. Tests and analysis of the obtained results allowed us to determine how additional layers of carbon nanostructures influenced on changes of static friction coefficients. At the same time, we analyzed the potential possibilities of applying nanostructures under consideration in mechanics.

Keywords: carbon nanotubes, static friction, dynamic friction

Procedia PDF Downloads 302

Authors: Yanqin Chen, Chao Jiang, Chongdu Cho

Abstract:

This research presents the three-dimensional mechanical characteristics of a commercial gas diffusion layer by experiment and simulation results. Although the mechanical performance of gas diffusion layers has attracted much attention, its reliability and accuracy are still a major challenge. With the help of simulation analysis methods, it is beneficial to the gas diffusion layer’s extensive commercial development and the overall stress analysis of proton electrolyte membrane fuel cells during its pre-production design period. Therefore, in this paper, a three-dimensional constitutive model of a commercial gas diffusion layer, including its material stiffness matrix parameters, is developed and coded, in the user-defined material model of a commercial finite element method software for simulation. Then, the model is validated by comparing experimental results as well as simulation outcomes. As a result, both the experimental data and simulation results show a good agreement with each other, with high accuracy.

Keywords: gas diffusion layer, proton electrolyte membrane fuel cell, stiffness matrix, three-dimensional mechanical characteristics, user-defined material model

Procedia PDF Downloads 146

Authors: Endang Jati Mat Sahid, Hussain Ali Bekhet

Abstract:

Ability to supply reliable and secure electricity has been one of the crucial components of economic development for any country. Forecasting of electricity production is therefore very important for accurate investment planning of generation power plants. In this study, we aim to examine and analyze the factors that affect electricity generation. Multiple regression models were used to find the relationship between various variables and electricity production. The models will simultaneously determine the effects of the variables on electricity generation. Many variables influencing electricity generation, i.e. natural gas (NG), coal (CO), fuel oil (FO), renewable energy (RE), gross domestic product (GDP) and fuel prices (FP), were examined for Malaysia. The results demonstrate that NG, CO, and FO were the main factors influencing electricity generation growth. This study then identified a number of policy implications resulting from the empirical results.

Keywords: energy policy, energy security, electricity production, Malaysia, the regression model

Procedia PDF Downloads 146

Authors: D. Subedi, S. Pradhan

Abstract:

Current transformers are an integral part of power system because it provides a proportional safe amount of current for protection and measurement applications. However when the power system experiences an abnormal situation leading to huge current flow, then this huge current is proportionally injected to the protection and metering circuit. Since the protection and metering equipment’s are designed to withstand only certain amount of current with respect to time, these high currents pose a risk to man and equipment. Therefore during such instances, the CT saturation characteristics have a huge influence on the safety of both man and equipment and also on the reliability of the protection and metering system. This paper shows the effect of burden on the Accuracy Limiting factor/ Instrument security factor of current transformers and also the change in saturation characteristics of the CT’s. The response of the CT to varying levels of overcurrent at different connected burden will be captured using the data acquisition software LabVIEW. Analysis is done on the real time data gathered using LabVIEW. Variation of current transformer saturation characteristics with changes in burden will be discussed.

Keywords: accuracy limiting factor, burden, current transformer, instrument security factor, saturation characteristics

Procedia PDF Downloads 405

Authors: Najoua El Hajjaji El Idrissi, Necip Gokhan Kasapoglu

Abstract:

Polarimetric Synthetic Aperture Radar-based imaging is a powerful technique used for earth observation and classification of surfaces. Forest evolution has been one of the vital areas of attention for the remote sensing experts. The information about forest areas can be achieved by remote sensing, whether by using active radars or optical instruments. However, due to several weather constraints, such as cloud cover, limited information can be recovered using optical data and for that reason, Polarimetric Synthetic Aperture Radar (PolSAR) is used as a powerful tool for forestry inventory. In this [14paper, we applied support vector machine (SVM) and Mahalanobis distance to the fully polarimetric AIRSAR P, L, C-bands data from the Nezer forest areas, the classification is based in the separation of different tree ages. The classification results were evaluated and the results show that the SVM performs better than the Mahalanobis distance and SVM achieves approximately 75% accuracy. This result proves that SVM classification can be used as a useful method to evaluate fully polarimetric SAR data with sufficient value of accuracy.

Keywords: classification, synthetic aperture radar, SAR polarimetry, support vector machine, mahalanobis distance

Procedia PDF Downloads 118

Authors: E. Omanya, E. Mueni, G. Makau, M. Kariuki

Abstract:

HIV infection is the strongest risk factor for a person to develop TB. Isoniazid preventive therapy (IPT) for People Living with HIV (PLWHIV) not only reduces the individual patients’ risk of developing active TB but mitigates cross infection. In Kenya, IPT for six months was recommended through the National TB, Leprosy and Lung Disease Program to treat latent TB. In spite of this recommendation by the national government, uptake of IPT among PLHIV remained low in Kenya by the end of 2015. The USAID/Kenya and East Africa Afya Jijini project, which supports 42 TBHIV health facilities in Nairobi County, began addressing low uptake of IPT through Quality Improvement (QI) teams set up at the facility level. Quality is characterized by WHO as one of the four main connectors between health systems building blocks and health systems outputs. Afya Jijini implements the Kenya Quality Model for Health, which involves QI teams being formed at the county, sub-county and facility levels. The teams review facility performance to identify gaps in service delivery and use QI tools to monitor and improve performance. Afya Jijini supported the formation of these teams in 42 facilities and built the teams’ capacity to review data and use QI principles to identify and address performance gaps. When the QI teams began working on improving IPT uptake among PLHIV, uptake was at 31.8%. The teams first conducted a root cause analysis using cause and effect diagrams, which help the teams to brainstorm on and to identify barriers to IPT uptake among PLHIV at the facility level. This is a participatory process where program staff provides technical support to the QI teams in problem identification and problem-solving. The gaps identified were inadequate knowledge and skills on the use of IPT among health care workers, lack of awareness of IPT by patients, inadequate monitoring and evaluation tools, and poor quantification and forecasting of IPT commodities. In response, Afya Jijini trained over 300 health care workers on the administration of IPT, supported patient education, supported quantification and forecasting of IPT commodities, and provided IPT data collection tools to help facilities monitor their performance. The facility QI teams conducted monthly meetings to monitor progress on implementation of IPT and took corrective action when necessary. IPT uptake improved from 31.8% to 61.2% during the second year of the Afya Jijini project and improved to 80.1% during the third year of the project’s support. Use of QI teams and root cause analysis to identify and address service delivery gaps, in addition to targeted program interventions and continual performance reviews, can be successful in increasing TB related service delivery uptake at health facilities.

Keywords: isoniazid, quality, health care workers, people leaving with HIV

Procedia PDF Downloads 89

Authors: Ruxandra Barbulescu, Daniel Ioan, Gabriela Ciuprina

Abstract:

The saltatory conduction is the way the action potential is transmitted along a myelinated axon. The potential diffuses along the myelinated compartments and it is regenerated in the Ranvier nodes due to the ion channels allowing the flow across the membrane. For an efficient simulation of populations of neurons, it is important to use reduced order models both for myelinated compartments and for Ranvier nodes and to have control over their accuracy and inner parameters. The paper presents a reduced order model of this neural system which allows an efficient simulation method for the saltatory conduction in myelinated axons. This model is obtained by concatenating reduced order linear models of 1D myelinated compartments and nonlinear 0D models of Ranvier nodes. The models for the myelinated compartments are selected from a series of spatially distributed models developed and hierarchized according to their modeling errors. The extracted model described by a nonlinear PDE of hyperbolic type is able to reproduce the saltatory conduction with acceptable accuracy and takes into account the finite propagation speed of potential. Finally, this model is again reduced in order to make it suitable for the inclusion in large-scale neural circuits.

Keywords: action potential, myelinated segments, nonlinear models, Ranvier nodes, reduced order models, saltatory conduction

Procedia PDF Downloads 144

Authors: Can Zhang, Qun Li, Yonglin Lei, Zhi Zhu, Dong Guo

Abstract:

Aiming at the problem of factor screen in distributed collaborative navigation of dense UAV swarm, an efficient distributed collaborative navigation factor screen method is proposed. The method considered the balance between computing load and positioning accuracy. The proposed algorithm utilized the factor graph model to implement a distributed collaborative navigation algorithm. The GNSS information of the UAV itself and the ranging information between the UAVs are used as the positioning factors. In this distributed scheme, a local factor graph is established for each UAV. The positioning factors of nodes with good geometric position distribution and small variance are selected to participate in the navigation calculation. To demonstrate and verify the proposed methods, the simulation and experiments in different scenarios are performed in this research. Simulation results show that the proposed scheme achieves a good balance between the computing load and positioning accuracy in the distributed cooperative navigation calculation of UAV swarm. This proposed algorithm has important theoretical and practical value for both industry and academic areas.

Keywords: screen method, cooperative positioning system, UAV swarm, factor graph, cooperative navigation

Procedia PDF Downloads 59