Search results for: semantic data

Authors: Zhen Cheng, Xinyu Dai, Shujian Huang, Jiajun Chen

Abstract:

Recently, explanatory natural language inference has attracted much attention for the interpretability of logic relationship prediction, which is also known as explanation generation for Natural Language Inference (NLI). Existing explanation generators based on discriminative Encoder-Decoder architecture have achieved noticeable results. However, we find that these discriminative generators usually generate explanations with correct evidence but incorrect logic semantic. It is due to that logic information is implicitly encoded in the premise-hypothesis pairs and difficult to model. Actually, logic information identically exists between premise-hypothesis pair and explanation. And it is easy to extract logic information that is explicitly contained in the target explanation. Hence we assume that there exists a latent space of logic information while generating explanations. Specifically, we propose a generative model called Variational Explanation Generator (VariationalEG) with a latent variable to model this space. Training with the guide of explicit logic information in target explanations, latent variable in VariationalEG could capture the implicit logic information in premise-hypothesis pairs effectively. Additionally, to tackle the problem of posterior collapse while training VariaztionalEG, we propose a simple yet effective approach called Logic Supervision on the latent variable to force it to encode logic information. Experiments on explanation generation benchmark—explanation-Stanford Natural Language Inference (e-SNLI) demonstrate that the proposed VariationalEG achieves significant improvement compared to previous studies and yields a state-of-the-art result. Furthermore, we perform the analysis of generated explanations to demonstrate the effect of the latent variable.

Keywords: natural language inference, explanation generation, variational auto-encoder, generative model

Procedia PDF Downloads 151

Authors: Takele Nemomsa, Girma Mamo, Tesfaye Balemi

Abstract:

Climate change refers to a change in the state of the climate that can be identified (e.g. using statistical tests) by changes in the mean and/or variance of its properties and that persists for an extended period, typically decades or longer. In Ethiopia; Maize production in relation to climate change at regional and sub- regional scales have not been studied in detail. Thus, this study was aimed to analyse the impact of climate change on maize yield in Ambo Districts, Central Ethiopia. To this effect, weather data, soil data and maize experimental data for Arganne hybrid were used. APSIM software was used to investigate the response of maize (Zea mays) yield to different agronomic management practices using current and future (2020s–2080s) climate data. The climate change projections data which were downscaled using SDSM were used as input of climate data for the impact analysis. Compared to agronomic practices the impact of climate change on Arganne in Central Ethiopia is minute. However, within 2020s-2080s in Ambo area; the yield of Arganne hybrid is projected to reduce by 1.06% to 2.02%, and in 2050s it is projected to reduce by 1.56 While in 2080s; it is projected to increase by 1.03% to 2.07%. Thus, to adapt to the changing climate; farmers should consider increasing plant density and fertilizer rate per hectare.

Keywords: APSIM, downscaling, response, SDSM

Procedia PDF Downloads 383

Authors: Rakesh Kumar, A. K. Ghosh

Abstract:

The paper presents the modeling of linear and nonlinear longitudinal aerodynamics using real flight data of Hansa-3 aircraft gathered at low and high angles of attack. The Neural-Gauss-Newton (NGN) method has been applied to model the linear and nonlinear longitudinal dynamics and estimate parameters from flight data. Unsteady aerodynamics due to flow separation at high angles of attack near stall has been included in the aerodynamic model using Kirchhoff’s quasi-steady stall model. NGN method is an algorithm that utilizes Feed Forward Neural Network (FFNN) and Gauss-Newton optimization to estimate the parameters and it does not require any a priori postulation of mathematical model or solving of equations of motion. NGN method was validated on real flight data generated at moderate angles of attack before application to the data at high angles of attack. The estimates obtained from compatible flight data using NGN method were validated by comparing with wind tunnel values and the maximum likelihood estimates. Validation was also carried out by comparing the response of measured motion variables with the response generated by using estimates a different control input. Next, NGN method was applied to real flight data generated by executing a well-designed quasi-steady stall maneuver. The results obtained in terms of stall characteristics and aerodynamic parameters were encouraging and reasonably accurate to establish NGN as a method for modeling nonlinear aerodynamics from real flight data at high angles of attack.

Keywords: parameter estimation, NGN method, linear and nonlinear, aerodynamic modeling

Procedia PDF Downloads 446

Authors: Gabriel Gomez

Abstract:

The very concept of information seeking behavior, and the means by which librarians teach users to gain information, that is information literacy, are at the heart of how libraries deliver information, but big data will forever change human interaction with information and the way such behavior is both studied and taught. Just as importantly, big data will orient the study of behavior towards commercial ends because of a tendency towards instrumentalist views of human behavior, something one might also call a trend towards behaviorism. This oral presentation seeks to explore how the impact of big data on understandings of human behavior might impact a library information science (LIS) view of human behavior and information literacy, and what this might mean for social justice aims and concomitant community action normally at the center of librarianship. The methodology employed here is a non-empirical examination of current understandings of LIS in regards to social justice alongside an examination of the benefits and dangers foreseen with the growth of big data analysis. The rise of big data within the ever-changing information environment encapsulates a shift to a more mechanistic view of human behavior, one that can easily encompass information seeking behavior and information use. As commercial aims displace the important political and ethical aims that are often central to the missions espoused by libraries and the social sciences, the very altruism and power relations found in LIS are at risk. In this oral presentation, an examination of the social justice impulses of librarians regarding power and information demonstrates how such impulses can be challenged by big data, particularly as librarians understand user behavior and promote information literacy. The creeping behaviorist impulse inherent in the emphasis big data places on specific solutions, that is answers to question that ask how, as opposed to larger questions that hint at an understanding of why people learn or use information threaten library information science ideals. Together with the commercial nature of most big data, this existential threat can harm the social justice nature of librarianship.

Keywords: big data, library information science, behaviorism, librarianship

Procedia PDF Downloads 383

Authors: Min Kyung An

Abstract:

In this paper, we study the data collection problem in Wireless Sensor Networks (WSNs) adopting the two interference models: The graph model and the more realistic physical interference model known as Signal-to-Interference-Noise-Ratio (SINR). The main issue of the problem is to compute schedules with the minimum number of timeslots, that is, to compute the minimum latency schedules, such that data from every node can be collected without any collision or interference to a sink node. While existing works studied the problem with unit-sized and unbounded-sized message models, we investigate the problem with the bounded-sized message model, and introduce a constant factor approximation algorithm. To the best known of our knowledge, our result is the first result of the data collection problem with bounded-sized model in both interference models.

Keywords: data collection, collision-free, interference-free, physical interference model, SINR, approximation, bounded-sized message model, wireless sensor networks

Procedia PDF Downloads 222

Authors: Chen Xi, Lao Xuerui, Li Junjie, Jiang Yike, Wang Hanwei, Zeng Zihao

Abstract:

To further promote the development of smart cities, the microscopic "nerve endings" of the City Intelligent Model (CIM) are extended to be more sensitive. In this paper, we develop a pedestrian trajectory twin terminal based on the CIM and CNN technology. It also uses 5G networks, architectural and geoinformatics technologies, convolutional neural networks, combined with deep learning networks for human behaviour recognition models, to provide empirical data such as 'pedestrian flow data and human behavioural characteristics data', and ultimately form spatial performance evaluation criteria and spatial performance warning systems, to make the empirical data accurate and intelligent for prediction and decision making.

Keywords: urban planning, urban governance, CIM, artificial intelligence, convolutional neural network

Procedia PDF Downloads 150

Authors: Yunos Zahri, Ariffin Aswami

Abstract:

Cyber terrorism represents the convergence of two worlds: virtual and physical. The virtual world is a place in which computer programs function and data move, whereas the physical world is where people live and function. The merging of these two domains is the interface being targeted in the incidence of cyber terrorism. To better understand why cyber terrorism acts are committed, this study presents the context of cyber terrorism from motivational perspectives. Motivational forces behind cyber terrorism can be social, political, ideological and economic. In this research, data are analyzed using a qualitative method. A semi-structured interview with purposive sampling was used for data collection. With the growing interconnectedness between critical infrastructures and Information & Communication Technology (ICT), selecting targets that facilitate maximum disruption can significantly influence terrorists. This work provides a baseline for defining the concept of cyber terrorism from motivational perspectives.

Keywords: cyber terrorism, terrorism, motivation, qualitative analysis

Procedia PDF Downloads 422

Authors: Sheheryar Khan, Weidong Li, Fanqian Meng

Abstract:

The Urban Heat Island (UHI) effect is one of the foremost problems out of other ecological and socioeconomic issues in urbanization. Due to this phenomenon that human-made urban areas have replaced the rural landscape with the surface that increases thermal conductivity and urban warmth; as a result, the temperature in the city is higher than in the surrounding rural areas. To affect the evidence of this phenomenon in the Zhengzhou city area, an observation of the temperature variations in the urban area is done through a scientific method that has been followed. Landsat 8 satellite images were taken from 2013 to 2015 to calculate the effect of Urban Heat Island (UHI) along with the NPP-VRRIS night-time remote sensing data to analyze the result for a better understanding of the center of the built-up area. To further support the evidence, the correlation between land surface temperatures and the normalized difference vegetation index (NDVI) was calculated using the Red band 4 and Near-infrared band 5 of the Landsat 8 data. Mono-window algorithm was applied to retrieve the land surface temperature (LST) distribution from the Landsat 8 data using Band 10 and 11 accordingly to convert the top-of-atmosphere radiance (TOA) and to convert the satellite brightness temperature. Along with Landsat 8 data, NPP-VIIRS night-light data is preprocessed to get the research area data. The analysis between Landsat 8 data and NPP night-light data was taken to compare the output center of the Built-up area of Zhengzhou city.

Keywords: built-up area, land surface temperature, mono-window algorithm, NDVI, remote sensing, threshold method, Zhengzhou

Procedia PDF Downloads 139

Authors: Robab Abdolkhani, Farzin Halabchi, Reza Safdari, Goli Arji

Abstract:

Background and purpose: The quality of health record depends on the quality of its content and proper documentation. Minimum data set makes a standard method for collecting key data elements that make them easy to understand and enable comparison. The aim of this study was to determine the minimum data set for Iranian athletes’ health records. Methods: This study is an applied research of a descriptive comparative type which was carried out in 2013. By using internal and external forms of documentation, a checklist was created that included data elements of athletes health record and was subjected to debate in Delphi method by experts in the field of sports medicine and health information management. Results: From 97 elements which were subjected to discussion, 85 elements by more than 75 percent of the participants (as the main elements) and 12 elements by 50 to 75 percent of the participants (as the proposed elements) were agreed upon. In about 97 elements of the case, there was no significant difference between responses of alumni groups of sport pathology and sports medicine specialists with medical record, medical informatics and information management professionals. Conclusion: Minimum data set of Iranian athletes’ health record with four information categories including demographic information, health history, assessment and treatment plan was presented. The proposed model is available for manual and electronic medical records.

Keywords: Documentation, Health record, Minimum data set, Sports medicine

Procedia PDF Downloads 480

Authors: Maria Antonieta Erna Castillo Holly

Abstract:

During the last decade, data analysis, strategic decision making, and the use of artificial intelligence (AI) tools in Latin American agriculture have been a challenge. In some countries, the availability, quality, and reliability of historical data, in addition to the current data recording methodology in the field, makes it difficult to use information systems, complete data analysis, and their support for making the right strategic decisions. This is something essential in Agriculture 4.0. where the increase in the global demand for fresh agricultural products of tropical origin, during all the seasons of the year requires a change in the production model and greater agility in the responses to the consumer market demands of quality, quantity, traceability, and sustainability –that means extensive data-. Having quality information available and updated in real-time on what, how much, how, when, where, at what cost, and the compliance with production quality standards represents the greatest challenge for sustainable and profitable agriculture in the region. The objective of this work is to present a methodological proposal for the collection of georeferenced data from the protected agriculture sector, specifically in production units (UP) with tall structures (Greenhouses), initially for Venezuela, taking the state of Mérida as the geographical framework, and horticultural products as target crops. The document presents some background information and explains the methodology and tools used in the 3 phases of the work: diagnosis, data collection, and analysis. As a result, an evaluation of the process is carried out, relevant data and dashboards are displayed, and the first satellite maps integrated with layers of information in a geographic information system are presented. Finally, some improvement proposals and tentatively recommended applications are added to the process, understanding that their objective is to provide better qualified and traceable georeferenced data for subsequent analysis of the information and more agile and accurate strategic decision making. One of the main points of this study is the lack of quality data treatment in the Latin America area and especially in the Caribbean basin, being one of the most important points how to manage the lack of complete official data. The methodology has been tested with horticultural products, but it can be extended to other tropical crops.

Keywords: greenhouses, protected agriculture, data analysis, geographic information systems, Venezuela

Procedia PDF Downloads 131

Authors: S. H. Lee, M. J. Park, O. M. Kwon

Abstract:

In this paper, reliable consensus of multi-agent systems with sampled-data is investigated. By using a suitable Lyapunov-Krasovskii functional and some techniques such as Wirtinger Inequality, Schur Complement and Kronecker Product, the results of this systems are obtained by solving a set of Linear Matrix Inequalities(LMIs). One numerical example is included to show the effectiveness of the proposed criteria.

Keywords: multi-agent, linear matrix inequalities (LMIs), kronecker product, sampled-data, Lyapunov method

Procedia PDF Downloads 528

Authors: Zahra Fotovatnia, Jeffery A. Jones, Alexandra Gottardo

Abstract:

Speech communication often occurs in environments where noise conceals part of a message. Listeners should compensate for the lack of auditory information by picking up distinct acoustic cues and using semantic and sentential context to recreate the speaker’s intended message. This situation seems to be more challenging in a nonnative than native language. On the other hand, early bilinguals are expected to show an advantage over the late bilingual and monolingual speakers of a language due to their better executive functioning components. In this study, English monolingual speakers were compared with early and late nonnative speakers of English to understand speech in noise processing (SIN) and the underlying neurobiological features of this phenomenon. Auditory mismatch negativities (MMNs) were recorded using a double-oddball paradigm in response to a minimal pair that differed in their middle vowel (beat/bit) at Wilfrid Laurier University in Ontario, Canada. The results did not show any significant structural and electroneural differences across groups. However, vocabulary knowledge correlated positively with performance on tests that measured SIN processing in participants who learned English after age 6. Moreover, their performance on the test negatively correlated with the integral area amplitudes in the left superior temporal gyrus (STG). In addition, the STG was engaged before the inferior frontal gyrus (IFG) in noise-free and low-noise test conditions in all groups. We infer that the pre-attentive processing of words engages temporal lobes earlier than the fronto-central areas and that vocabulary knowledge helps the nonnative perception of degraded speech.

Keywords: degraded speech perception, event-related brain potentials, mismatch negativities, brain regions

Procedia PDF Downloads 107

Authors: Yusuf Ziya Ayık, Ferhat Kahveci

Abstract:

Currently, database management systems have various tools such as backup and maintenance, and also provide statistical information such as resource usage and security. In terms of query performance, this paper covers query optimization, views, indexed tables, pre-computation materialized view, query performance analysis in which query plan alternatives can be created and the least costly one selected to optimize a query. Indexes and views can be created for related table columns. The literature review of this study showed that, in the course of time, despite the growing capabilities of the database management system, only database administrators are aware of the need for dealing with archival and transactional data types differently. These data may be constantly changing data used in everyday life, and also may be from the completed questionnaire whose data input was completed. For both types of data, the database uses its capabilities; but as shown in the findings section, instead of repeating similar heavy calculations which are carrying out same results with the same query over a survey results, using materialized view results can be in a more simple way. In this study, this performance difference was observed quantitatively considering the cost of the query.

Keywords: cost of query, database management systems, materialized view, query performance

Procedia PDF Downloads 280

Authors: Chia-Hau Liu, Tai-Yue Wang

Abstract:

Traditional multivariate control charts assume that measurement from manufacturing processes follows a multivariate normal distribution. However, this assumption may not hold or may be difficult to verify because not all the measurement from manufacturing processes are normal distributed in practice. This study develops a new multivariate control chart for monitoring the processes with non-normal data. We propose a mechanism based on integrating the one-class classification method and the adaptive technique. The adaptive technique is used to improve the sensitivity to small shift on one-class classification in statistical process control. In addition, this design provides an easy way to allocate the value of type I error so it is easier to be implemented. Finally, the simulation study and the real data from industry are used to demonstrate the effectiveness of the propose control charts.

Keywords: multivariate control chart, statistical process control, one-class classification method, non-normal data

Procedia PDF Downloads 423

Authors: Samuel Munaf, Kevin Swingler, Franz Brülisauer, Anthony O’Hare, George Gunn, Aaron Reeves

Abstract:

Web scraping and text mining are popular computer science methods deployed by public health researchers to augment traditional epidemiological surveillance. However, within veterinary disease surveillance, such techniques are still in the early stages of development and have not yet been fully utilised. This study presents an exploration into the utility of incorporating internet-based data to better understand the smallholder farming communities within Scotland by using online text extraction and the subsequent mining of this data. Web scraping of the livestock fora was conducted in conjunction with text mining of the data in search of common themes, words, and topics found within the text. Results from bi-grams and topic modelling uncover four main topics of interest within the data pertaining to aspects of livestock husbandry: feeding, breeding, slaughter, and disposal. These topics were found amongst both the poultry and pig sub-forums. Topic modeling appears to be a useful method of unsupervised classification regarding this form of data, as it has produced clusters that relate to biosecurity and animal welfare. Internet data can be a very effective tool in aiding traditional veterinary surveillance methods, but the requirement for human validation of said data is crucial. This opens avenues of research via the incorporation of other dynamic social media data, namely Twitter and Facebook/Meta, in addition to time series analysis to highlight temporal patterns.

Keywords: veterinary epidemiology, disease surveillance, infodemiology, infoveillance, smallholding, social media, web scraping, sentiment analysis, geolocation, text mining, NLP

Procedia PDF Downloads 99

Authors: M. Koray Cetin, Mehmet Mert

Abstract:

The purpose of the study is to examine impacts on tourism revenues of the exchange rate and country overall security level. There are numerous studies that examine the bidirectional relation between macroeconomic factors and tourism revenues and tourism demand. Most of the studies support the existence of impact of tourism revenues on growth rate but not vice versa. Few studies examine the impact of factors like real exchange rate or purchasing power parity on the tourism revenues. In this context, firstly impact of real exchange rate on tourism revenues examination is aimed. Because exchange rate is one of the main determinants of international tourism services price in guests currency unit. Another determinant of tourism demand for a country is country’s overall security level. This issue can be handled in the context of the relationship between tourism revenues and overall security including turmoil, terrorism, border problem, political violence. In this study, factors are handled for several countries which have tourism revenues on a certain level. With this structure, it is a panel data, and it is evaluated with panel data analysis techniques. Panel data have at least two dimensions, and one of them is time dimensions. The panel data analysis techniques are applied to data gathered from Worldbank data web page. In this study, it is expected to find impacts of real exchange rate and security factors on tourism revenues for the countries that have noteworthy tourism revenues.

Keywords: exchange rate, panel data analysis, security, tourism revenues

Procedia PDF Downloads 351

Authors: Sumedha Ganjoo, Santosh Goswami

Abstract:

The rising reliance on technology places national security at the forefront of 21st-century issues. It complicates the efforts of emerging and developed countries to combat cyber threats and increases the inherent risk factors connected with technology. The inability to preserve data securely might have devastating repercussions on a massive scale. Consequently, it is vital to establish national, regional, and global data protection rules and regulations that penalise individuals who participate in immoral technology usage and exploit the inherent vulnerabilities of technology. This study paper seeks to analyse GDPR-inspired Bills in the South Asian Region and determine their suitability for the development of a worldwide data protection framework, considering that Asian countries are much more diversified than European ones. In light of this context, the objectives of this paper are to identify GDPR-inspired Bills in the South Asian Region, identify their similarities and differences, as well as the obstacles to developing a regional-level data protection mechanism, thereby satisfying the need to develop a global-level mechanism. Due to the qualitative character of this study, the researcher did a comprehensive literature review of prior research papers, journal articles, survey reports, and government publications on the aforementioned topics. Taking into consideration the survey results, the researcher conducted a critical analysis of the significant parameters highlighted in the literature study. Many nations in the South Asian area are in the process of revising their present data protection measures in accordance with GDPR, according to the primary results of this study. Consideration is given to the data protection laws of Thailand, Malaysia, China, and Japan. Significant parallels and differences in comparison to GDPR have been discussed in detail. The conclusion of the research analyses the development of various data protection legislation regimes in South Asia.

Keywords: data privacy, GDPR, Asia, data protection laws

Procedia PDF Downloads 82

Authors: Musab Isah

Abstract:

This paper presents a longitudinal analysis of Internet speed data in the Gulf Cooperation Council (GCC) region, focusing on the most populous cities of each of the six countries – Riyadh, Saudi Arabia; Dubai, UAE; Kuwait City, Kuwait; Doha, Qatar; Manama, Bahrain; and Muscat, Oman. The study utilizes data collected from the Measurement Lab (M-Lab) infrastructure over a five-year period from January 1, 2019, to December 31, 2023. The analysis includes downstream and upstream throughput data for the cities, covering significant events such as the launch of 5G networks in 2019, COVID-19-induced lockdowns in 2020 and 2021, and the subsequent recovery period and return to normalcy. The results showcase substantial increases in Internet speeds across the cities, highlighting improvements in both download and upload throughput over the years. All the GCC countries have achieved above-average Internet speeds that can conveniently support various online activities and applications with excellent user experience.

Keywords: internet data science, internet performance measurement, throughput analysis, internet speed, measurement lab, network diagnostic tool

Procedia PDF Downloads 63

Authors: Rose A. Yemson, Ping Jiang, Oyedeji L. Inumoh

Abstract:

The deployment of wireless sensor network has rapidly increased, however with the increased capacity and diversity of sensors, and applications ranging from biological, environmental, military etc. generates tremendous volume of data’s where more attention is placed on the distributed sensing and little on how to manage, analyze, retrieve and understand the data generated. This makes it more quite difficult to process live sensor data, run concurrent control and update because sensor data are either heavyweight, complex, and slow. This work will focus on developing a web service platform for automatic detection of sensors, acquisition of sensor data, storage of sensor data into a database, processing of sensor data using reconfigurable software components. This work will also create a web service based sensor data management system to monitor physical movement of an individual wearing wireless network sensor technology (SunSPOT). The sensor will detect movement of that individual by sensing the acceleration in the direction of X, Y and Z axes accordingly and then send the sensed reading to a database that will be interfaced with an internet platform. The collected sensed data will determine the posture of the person such as standing, sitting and lying down. The system is designed using the Unified Modeling Language (UML) and implemented using Java, JavaScript, html and MySQL. This system allows real time monitoring an individual closely and obtain their physical activity details without been physically presence for in-situ measurement which enables you to work remotely instead of the time consuming check of an individual. These details can help in evaluating an individual’s physical activity and generate feedback on medication. It can also help in keeping track of any mandatory physical activities required to be done by the individuals. These evaluations and feedback can help in maintaining a better health status of the individual and providing improved health care.

Keywords: HTML, java, javascript, MySQL, sunspot, UML, web-based, wireless network sensor

Procedia PDF Downloads 212

Authors: Valentina Marutyan

Abstract:

Healthcare data mining is a rapidly developing field at the intersection of technology and medicine that has the potential to change our understanding and approach to providing healthcare. Healthcare and data mining is the process of examining huge amounts of data to extract useful information that can be applied in order to improve patient care, treatment effectiveness, and overall healthcare delivery. This field looks for patterns, trends, and correlations in a variety of healthcare datasets, such as electronic health records (EHRs), medical imaging, patient demographics, and treatment histories. To accomplish this, it uses advanced analytical approaches. Predictive analysis using historical patient data is a major area of interest in healthcare data mining. This enables doctors to get involved early to prevent problems or improve results for patients. It also assists in early disease detection and customized treatment planning for every person. Doctors can customize a patient's care by looking at their medical history, genetic profile, current and previous therapies. In this way, treatments can be more effective and have fewer negative consequences. Moreover, helping patients, it improves the efficiency of hospitals. It helps them determine the number of beds or doctors they require in regard to the number of patients they expect. In this project are used models like logistic regression, random forests, and neural networks for predicting diseases and analyzing medical images. Patients were helped by algorithms such as k-means, and connections between treatments and patient responses were identified by association rule mining. Time series techniques helped in resource management by predicting patient admissions. These methods improved healthcare decision-making and personalized treatment. Also, healthcare data mining must deal with difficulties such as bad data quality, privacy challenges, managing large and complicated datasets, ensuring the reliability of models, managing biases, limited data sharing, and regulatory compliance. Finally, secret code of data mining in healthcare helps medical professionals and hospitals make better decisions, treat patients more efficiently, and work more efficiently. It ultimately comes down to using data to improve treatment, make better choices, and simplify hospital operations for all patients.

Keywords: data mining, healthcare, big data, large amounts of data

Procedia PDF Downloads 76

Authors: Zhao Gao, Eran Edirisinghe

Abstract:

The production of an accurate sketch of a suspect based on a verbal description obtained from a witness is an essential task for most criminal investigations. The criminal investigation system employs specifically trained professional artists to manually draw a facial image of the suspect according to the descriptions of an eyewitness for subsequent identification. Within the advancement of Deep Learning, Recurrent Neural Networks (RNN) have shown great promise in Natural Language Processing (NLP) tasks. Additionally, Generative Adversarial Networks (GAN) have also proven to be very effective in image generation. In this study, a trained GAN conditioned on textual features such as keywords automatically encoded from a verbal description of a human face using an RNN is used to generate photo-realistic facial images for criminal investigations. The intention of the proposed system is to map corresponding features into text generated from verbal descriptions. With this, it becomes possible to generate many reasonably accurate alternatives to which the witness can use to hopefully identify a suspect from. This reduces subjectivity in decision making both by the eyewitness and the artist while giving an opportunity for the witness to evaluate and reconsider decisions. Furthermore, the proposed approach benefits law enforcement agencies by reducing the time taken to physically draw each potential sketch, thus increasing response times and mitigating potentially malicious human intervention. With publically available 'CelebFaces Attributes Dataset' (CelebA) and additionally providing verbal description as training data, the proposed architecture is able to effectively produce facial structures from given text. Word Embeddings are learnt by applying the RNN architecture in order to perform semantic parsing, the output of which is fed into the GAN for synthesizing photo-realistic images. Rather than the grid search method, a metaheuristic search based on genetic algorithms is applied to evolve the network with the intent of achieving optimal hyperparameters in a fraction the time of a typical brute force approach. With the exception of the ‘CelebA’ training database, further novel test cases are supplied to the network for evaluation. Witness reports detailing criminals from Interpol or other law enforcement agencies are sampled on the network. Using the descriptions provided, samples are generated and compared with the ground truth images of a criminal in order to calculate the similarities. Two factors are used for performance evaluation: The Structural Similarity Index (SSIM) and the Peak Signal-to-Noise Ratio (PSNR). A high percentile output from this performance matrix should attribute to demonstrating the accuracy, in hope of proving that the proposed approach can be an effective tool for law enforcement agencies. The proposed approach to criminal facial image generation has potential to increase the ratio of criminal cases that can be ultimately resolved using eyewitness information gathering.

Keywords: RNN, GAN, NLP, facial composition, criminal investigation

Procedia PDF Downloads 162

Authors: Bushra Zafar, Usman Qamar

Abstract:

Large data sample size and dimensions render the effectiveness of conventional data mining methodologies. A data mining technique are important tools for collection of knowledgeable information from variety of databases and provides supervised learning in the form of classification to design models to describe vital data classes while structure of the classifier is based on class attribute. Classification efficiency and accuracy are often influenced to great extent by noisy and undesirable features in real application data sets. The inherent natures of data set greatly masks its quality analysis and leave us with quite few practical approaches to use. To our knowledge first time, we present a new approach for investigation of structure and quality of datasets by providing a targeted analysis of localization of noisy and irrelevant features of data sets. Machine learning is based primarily on feature selection as pre-processing step which offers us to select few features from number of features as a subset by reducing the space according to certain evaluation criterion. The primary objective of this study is to trim down the scope of the given data sample by searching a small set of important features which may results into good classification performance. For this purpose, a heuristic for wrapper-based feature selection using genetic algorithm and for discriminative feature selection an external classifier are used. Selection of feature based on its number of occurrence in the chosen chromosomes. Sample dataset has been used to demonstrate proposed idea effectively. A proposed method has improved average accuracy of different datasets is about 95%. Experimental results illustrate that proposed algorithm increases the accuracy of prediction of different diseases.

Keywords: data mining, generic algorithm, KNN algorithms, wrapper based feature selection

Procedia PDF Downloads 316

Authors: Wade Ghribi, Abdelmoty M. Ahmed, Ahmed Said Badawy, Belgacem Bouallegue

Abstract:

In higher education institutions, the most pressing priority is to improve student performance and retention. Large volumes of student data are used in Educational Data Mining techniques to find new hidden information from students' learning behavior, particularly to uncover the early symptom of at-risk pupils. On the other hand, data with noise, outliers, and irrelevant information may provide incorrect conclusions. By identifying features of students' data that have the potential to improve performance prediction results, comparing and identifying the most appropriate ensemble learning technique after preprocessing the data, and optimizing the hyperparameters, this paper aims to develop a reliable students' performance prediction model for Higher Education Institutions. Data was gathered from two different systems: a student information system and an e-learning system for undergraduate students in the College of Computer Science of a Saudi Arabian State University. The cases of 4413 students were used in this article. The process includes data collection, data integration, data preprocessing (such as cleaning, normalization, and transformation), feature selection, pattern extraction, and, finally, model optimization and assessment. Random Forest, Bagging, Stacking, Majority Vote, and two types of Boosting techniques, AdaBoost and XGBoost, are ensemble learning approaches, whereas Decision Tree, Support Vector Machine, and Artificial Neural Network are supervised learning techniques. Hyperparameters for ensemble learning systems will be fine-tuned to provide enhanced performance and optimal output. The findings imply that combining features of students' behavior from e-learning and students' information systems using Majority Vote produced better outcomes than the other ensemble techniques.

Keywords: educational data mining, student performance prediction, e-learning, classification, ensemble learning, higher education

Procedia PDF Downloads 108

Authors: Hae-Won Seo, Yong-Gu Lee

Abstract:

Recent progress in the next generation of automobile technology is geared towards incorporating information technology into cars. Collectively called smart cars are bringing intelligence to cars that provides comfort, convenience and safety. A branch of smart cars is connected-car system. The key concept in connected-cars is the sharing of driving information among cars through decentralized manner enabling collective intelligence. This paper proposes a foundation of the information model that is necessary to define the driving information for smart-cars. Road conditions are modeled through a unique data structure that unambiguously represent the time variant traffics in the streets. Additionally, the modeled data structure is exemplified in a navigational scenario and usage using UML. Optimal driving route searching is also discussed using the proposed data structure in a dynamically changing road conditions.

Keywords: connected-car, data modeling, route planning, navigation system

Procedia PDF Downloads 374

Authors: Georgii Emelianov, Mikhail Polikarpov, Fabian Hübner, Jochen Deuse, Jochen Schiemann

Abstract:

Recycling refrigerating appliances plays a major role in protecting the Earth's atmosphere from ozone depletion and emissions of greenhouse gases. The performance of refrigerator recycling plants in terms of material retention is the subject of strict environmental certifications and is reviewed periodically through specialized audits. The continuous collection of Refrigerator data required for the input-output analysis is still mostly manual, error-prone, and not digitalized. In this paper, we propose an automated data collection system for recycling plants in order to deduce expected material contents in individual end-of-life refrigerating appliances. The system utilizes laser scanner measurements and optical data to extract attributes of individual refrigerators by applying transfer learning with pre-trained vision models and optical character recognition. Based on Recognized features, the system automatically provides material categories and target values of contained material masses, especially foaming and cooling agents. The presented data collection system paves the way for continuous performance monitoring and efficient control of refrigerator recycling plants.

Keywords: automation, data collection, performance monitoring, recycling, refrigerators

Procedia PDF Downloads 164

Authors: Soojin Kim, Jiwon Yang, Sungzoon Cho

Abstract:

As a seasonal product is only in demand for a short time, inventory management is critical to profits. Both markdowns and stockouts decrease the return on perishable products; therefore, researchers have been interested in the distribution of seasonal products with the aim of maximizing profits. In this study, we propose a data-driven seasonal product sales pattern analysis method for individual retail outlets based on observed sales data clustering; the proposed method helps in determining distribution strategies.

Keywords: clustering, distribution, sales pattern, seasonal product

Procedia PDF Downloads 597

Authors: Surya R. Niraula, Devendra B Chhetry, Girish K. Singh, S. Nagesh, Frederick A. Connell

Abstract:

Background: Although random sampling is generally considered to be the gold standard for population-based research, the majority of drug abuse research is based on non-random sampling despite the well-known limitations of this kind of sampling. Method: We compared the statistical properties of two surveys of drug abuse in the same community: one using snowball sampling of drug users who then identified “friend controls” and the other using a random sample of non-drug users (controls) who then identified “friend cases.” Models to predict drug abuse based on risk factors were developed for each data set using conditional logistic regression. We compared the precision of each model using bootstrapping method and the predictive properties of each model using receiver operating characteristics (ROC) curves. Results: Analysis of 100 random bootstrap samples drawn from the snowball-sample data set showed a wide variation in the standard errors of the beta coefficients of the predictive model, none of which achieved statistical significance. One the other hand, bootstrap analysis of the random-sample data set showed less variation, and did not change the significance of the predictors at the 5% level when compared to the non-bootstrap analysis. Comparison of the area under the ROC curves using the model derived from the random-sample data set was similar when fitted to either data set (0.93, for random-sample data vs. 0.91 for snowball-sample data, p=0.35); however, when the model derived from the snowball-sample data set was fitted to each of the data sets, the areas under the curve were significantly different (0.98 vs. 0.83, p < .001). Conclusion: The proposed method of random sampling of controls appears to be superior from a statistical perspective to snowball sampling and may represent a viable alternative to snowball sampling.

Keywords: drug abuse, matched case-control study, non-probability sampling, probability sampling

Procedia PDF Downloads 493

Authors: Javed Mohammed

Abstract:

Right now, bio-medical infrastructure lags well behind the curve. Our healthcare system is dispersed and disjointed; medical records are a bit of a mess; and we do not yet have the capacity to store and process the crazy amounts of data coming our way from widespread whole-genome sequencing. And then there are privacy issues. Despite these infrastructure challenges, some researchers are plunging into bio medical Big Data now, in hopes of extracting new and actionable knowledge. They are doing delving into molecular-level data to discover bio markers that help classify patients based on their response to existing treatments; and pushing their results out to physicians in novel and creative ways. Computer scientists and bio medical researchers are able to transform data into models and simulations that will enable scientists for the first time to gain a profound under-standing of the deepest biological functions. Solving biological problems may require High-Performance Computing HPC due either to the massive parallel computation required to solve a particular problem or to algorithmic complexity that may range from difficult to intractable. Many problems involve seemingly well-behaved polynomial time algorithms (such as all-to-all comparisons) but have massive computational requirements due to the large data sets that must be analyzed. High-throughput techniques for DNA sequencing and analysis of gene expression have led to exponential growth in the amount of publicly available genomic data. With the increased availability of genomic data traditional database approaches are no longer sufficient for rapidly performing life science queries involving the fusion of data types. Computing systems are now so powerful it is possible for researchers to consider modeling the folding of a protein or even the simulation of an entire human body. This research paper emphasizes the computational biology's growing need for high-performance computing and Big Data. It illustrates this article’s indispensability in meeting the scientific and engineering challenges of the twenty-first century, and how Protein Folding (the structure and function of proteins) and Phylogeny Reconstruction (evolutionary history of a group of genes) can use HPC that provides sufficient capability for evaluating or solving more limited but meaningful instances. This article also indicates solutions to optimization problems, and benefits Big Data and Computational Biology. The article illustrates the Current State-of-the-Art and Future-Generation Biology of HPC Computing with Big Data.

Keywords: high performance, big data, parallel computation, molecular data, computational biology

Procedia PDF Downloads 364

Authors: Natividad Natalia Angeles Manuel

Abstract:

In the Dominican Republic, American-Dominican veterans face unique challenges due to their dual identities and wartime experiences. This case study examines an 85-year-old veteran with memory impairments and emotional distress linked to military service. A neuropsychological assessment using standardized tools evaluated cognitive domains and functional abilities. Significant deficits in memory, orientation, semantic memory, and executive functions, alongside symptoms of Post-Traumatic Stress Disorder and depression, were identified. A six-month cognitive stimulation program included tailored interventions to enhance memory, attention, and executive skills through weekly sessions and group activities. Medical and physical therapy support aimed to improve overall cognitive, functional, and emotional outcomes. Follow-up evaluations showed improvements in memory retention, attention, task proficiency, and reduced depressive symptoms, highlighting the program's effectiveness in promoting emotional well-being and quality of life. Despite ongoing memory challenges and military-related nightmares, the veteran responded positively to interventions, demonstrating resilience and motivation. This study emphasizes the importance of personalized neuropsychological interventions for American-Dominican veterans in the Dominican Republic. Through assessment tools and focused cognitive stimulation strategies, healthcare providers can successfully alleviate cognitive and emotional challenges stemming from traumatic experiences in elderly veterans. Overall, integrated neuropsychological assessment and stimulation programs are shown to enhance cognitive resilience and emotional well-being, thus contributing to an enhanced quality of life for aging American-Dominican veterans.

Keywords: neuropsychology, cognitive stimulation, American-Dominican veterans, Dominican Republic, PTSD, memory deficits

Procedia PDF Downloads 36

Authors: Durjoy Lahiri, Vishal Madhukar Sawale, Ashwani Bhat, Souvik Dubey, Gautam Das, Biman Kanti Roy, Suparna Chatterjee, Goutam Gangopadhyay

Abstract:

Crossed aphasia has been an area of considerable interest for cognitive researchers as it offers a fascinating insight into cerebral lateralization for language function. We conducted an observational study in the stroke unit of a tertiary care neurology teaching hospital in eastern India on subjects with crossed aphasia over a period of four years. During the study period, we detected twelve cases of crossed aphasia in strongly right-handed patients, caused by ischemic stroke. The age, gender, vernacular language and educational status of the patients were noted. Aphasia type and severity were assessed using Bengali version of Western Aphasia Battery (validated). Computed tomography, magnetic resonance imaging and angiography were used to evaluate the location and extent of the ischemic lesion in brain. Our series of 12 cases of crossed aphasia included 7 male and 5 female with mean age being 58.6 years. Eight patients were found to have Broca’s aphasia, 3 had trans-cortical motor aphasia and 1 patient suffered from global aphasia. Nine patients were having very severe aphasia and 3 suffered from mild aphasia. Mirror-image type of crossed aphasia was found in 3 patients, whereas 9 had anomalous variety. In our study crossed aphasia was found to be more frequent in males. Anomalous pattern was more common than mirror-image. Majority of the patients had motor-type aphasia and no patient was found to have pure comprehension deficit. We hypothesize that in Bengali-speaking right-handed population, lexical-semantic system of the language network remains loyal to the left hemisphere even if the phonological output system is anomalously located in the right hemisphere.

Keywords: aphasia, crossed, lateralization, language function, vascular

Procedia PDF Downloads 192