Search results for: wireless senor networks
1286 Embedded Semantic Segmentation Network Optimized for Matrix Multiplication Accelerator
Authors: Jaeyoung Lee
Abstract:
Autonomous driving systems require high reliability to provide people with a safe and comfortable driving experience. However, despite the development of a number of vehicle sensors, it is difficult to always provide high perceived performance in driving environments that vary from time to season. The image segmentation method using deep learning, which has recently evolved rapidly, provides high recognition performance in various road environments stably. However, since the system controls a vehicle in real time, a highly complex deep learning network cannot be used due to time and memory constraints. Moreover, efficient networks are optimized for GPU environments, which degrade performance in embedded processor environments equipped simple hardware accelerators. In this paper, a semantic segmentation network, matrix multiplication accelerator network (MMANet), optimized for matrix multiplication accelerator (MMA) on Texas instrument digital signal processors (TI DSP) is proposed to improve the recognition performance of autonomous driving system. The proposed method is designed to maximize the number of layers that can be performed in a limited time to provide reliable driving environment information in real time. First, the number of channels in the activation map is fixed to fit the structure of MMA. By increasing the number of parallel branches, the lack of information caused by fixing the number of channels is resolved. Second, an efficient convolution is selected depending on the size of the activation. Since MMA is a fixed, it may be more efficient for normal convolution than depthwise separable convolution depending on memory access overhead. Thus, a convolution type is decided according to output stride to increase network depth. In addition, memory access time is minimized by processing operations only in L3 cache. Lastly, reliable contexts are extracted using the extended atrous spatial pyramid pooling (ASPP). The suggested method gets stable features from an extended path by increasing the kernel size and accessing consecutive data. In addition, it consists of two ASPPs to obtain high quality contexts using the restored shape without global average pooling paths since the layer uses MMA as a simple adder. To verify the proposed method, an experiment is conducted using perfsim, a timing simulator, and the Cityscapes validation sets. The proposed network can process an image with 640 x 480 resolution for 6.67 ms, so six cameras can be used to identify the surroundings of the vehicle as 20 frame per second (FPS). In addition, it achieves 73.1% mean intersection over union (mIoU) which is the highest recognition rate among embedded networks on the Cityscapes validation set.Keywords: edge network, embedded network, MMA, matrix multiplication accelerator, semantic segmentation network
Procedia PDF Downloads 1181285 Reconfigurable Ubiquitous Computing Infrastructure for Load Balancing
Authors: Khaled Sellami, Lynda Sellami, Pierre F. Tiako
Abstract:
Ubiquitous computing helps make data and services available to users anytime and anywhere. This makes the cooperation of devices a crucial need. In return, such cooperation causes an overload of the devices and/or networks, resulting in network malfunction and suspension of its activities. Our goal in this paper is to propose an approach of devices reconfiguration in order to help to reduce the energy consumption in ubiquitous environments. The idea is that when high-energy consumption is detected, we proceed to a change in component distribution on the devices to reduce and/or balance the energy consumption. We also investigate the possibility to detect high-energy consumption of devices/network based on devices abilities. As a result, our idea realizes a reconfiguration of devices aimed at reducing the consumption of energy and/or load balancing in ubiquitous environments.Keywords: ubiquitous computing, load balancing, device energy consumption, reconfiguration
Procedia PDF Downloads 2661284 Detection and Quantification of Ochratoxin A in Food by Aptasensor
Authors: Moez Elsaadani, Noel Durand, Brice Sorli, Didier Montet
Abstract:
Governments and international instances are trying to improve the food safety system to prevent, reduce or avoid the increase of food borne diseases. This food risk is one of the major concerns for the humanity. The contamination by mycotoxins is a threat to the health and life of humans and animals. One of the most common mycotoxin contaminating feed and foodstuffs is Ochratoxin A (OTA), which is a secondary metabolite, produced by Aspergillus and Penicillium strains. OTA has a chronic toxic effect and proved to be mutagenic, nephrotoxic, teratogenic, immunosuppressive, and carcinogenic. On the other side, because of their high stability, specificity, affinity, and their easy chemical synthesis, aptamer based methods are applied to OTA biosensing as alternative to traditional analytical technique. In this work, five aptamers have been tested to confirm qualitatively and quantitatively their binding with OTA. In the same time, three different analytical methods were tested and compared based on their ability to detect and quantify the OTA. The best protocol that was established to quantify free OTA from linked OTA involved an ultrafiltration method in green coffee solution with. OTA was quantified by HPLC-FLD to calculate the binding percentage of all five aptamers. One aptamer (The most effective with 87% binding with OTA) has been selected to be our biorecognition element to study its electrical response (variation of electrical properties) in the presence of OTA in order to be able to make a pairing with a radio frequency identification (RFID). This device, which is characterized by its low cost, speed, and a simple wireless information transmission, will implement the knowledge on the mycotoxins molecular sensors (aptamers), an electronic device that will link the information, the quantification and make it available to operators.Keywords: aptamer, aptasensor, detection, Ochratoxin A
Procedia PDF Downloads 1661283 The Location-Routing Problem with Pickup Facilities and Heterogeneous Demand: Formulation and Heuristics Approach
Authors: Mao Zhaofang, Xu Yida, Fang Kan, Fu Enyuan, Zhao Zhao
Abstract:
Nowadays, last-mile distribution plays an increasingly important role in the whole industrial chain delivery link and accounts for a large proportion of the whole distribution process cost. Promoting the upgrading of logistics networks and improving the layout of final distribution points has become one of the trends in the development of modern logistics. Due to the discrete and heterogeneous needs and spatial distribution of customer demand, which will lead to a higher delivery failure rate and lower vehicle utilization, last-mile delivery has become a time-consuming and uncertain process. As a result, courier companies have introduced a range of innovative parcel storage facilities, including pick-up points and lockers. The introduction of pick-up points and lockers has not only improved the users’ experience but has also helped logistics and courier companies achieve large-scale economy. Against the backdrop of the COVID-19 of the previous period, contactless delivery has become a new hotspot, which has also created new opportunities for the development of collection services. Therefore, a key issue for logistics companies is how to design/redesign their last-mile distribution network systems to create integrated logistics and distribution networks that consider pick-up points and lockers. This paper focuses on the introduction of self-pickup facilities in new logistics and distribution scenarios and the heterogeneous demands of customers. In this paper, we consider two types of demand, including ordinary products and refrigerated products, as well as corresponding transportation vehicles. We consider the constraints associated with self-pickup points and lockers and then address the location-routing problem with self-pickup facilities and heterogeneous demands (LRP-PFHD). To solve this challenging problem, we propose a mixed integer linear programming (MILP) model that aims to minimize the total cost, which includes the facility opening cost, the variable transport cost, and the fixed transport cost. Due to the NP-hardness of the problem, we propose a hybrid adaptive large-neighbourhood search algorithm to solve LRP-PFHD. We evaluate the effectiveness and efficiency of the proposed algorithm by using instances generated based on benchmark instances. The results demonstrate that the hybrid adaptive large neighbourhood search algorithm is more efficient than MILP solvers such as Gurobi for LRP-PFHD, especially for large-scale instances. In addition, we made a comprehensive analysis of some important parameters (e.g., facility opening cost and transportation cost) to explore their impacts on the results and suggested helpful managerial insights for courier companies.Keywords: city logistics, last-mile delivery, location-routing, adaptive large neighborhood search
Procedia PDF Downloads 681282 Foot Recognition Using Deep Learning for Knee Rehabilitation
Authors: Rakkrit Duangsoithong, Jermphiphut Jaruenpunyasak, Alba Garcia
Abstract:
The use of foot recognition can be applied in many medical fields such as the gait pattern analysis and the knee exercises of patients in rehabilitation. Generally, a camera-based foot recognition system is intended to capture a patient image in a controlled room and background to recognize the foot in the limited views. However, this system can be inconvenient to monitor the knee exercises at home. In order to overcome these problems, this paper proposes to use the deep learning method using Convolutional Neural Networks (CNNs) for foot recognition. The results are compared with the traditional classification method using LBP and HOG features with kNN and SVM classifiers. According to the results, deep learning method provides better accuracy but with higher complexity to recognize the foot images from online databases than the traditional classification method.Keywords: foot recognition, deep learning, knee rehabilitation, convolutional neural network
Procedia PDF Downloads 1511281 Automatic Measurement of Garment Sizes Using Deep Learning
Authors: Maulik Parmar, Sumeet Sandhu
Abstract:
The online fashion industry experiences high product return rates. Many returns are because of size/fit mismatches -the size scale on labels can vary across brands, the size parameters may not capture all fit measurements, or the product may have manufacturing defects. Warehouse quality check of garment sizes can be semi-automated to improve speed and accuracy. This paper presents an approach for automatically measuring garment sizes from a single image of the garment -using Deep Learning to learn garment keypoints. The paper focuses on the waist size measurement of jeans and can be easily extended to other garment types and measurements. Experimental results show that this approach can greatly improve the speed and accuracy of today’s manual measurement process.Keywords: convolutional neural networks, deep learning, distortion, garment measurements, image warping, keypoints
Procedia PDF Downloads 2891280 Performance Evaluation of Routing Protocol in Cognitive Radio with Multi Technological Environment
Authors: M. Yosra, A. Mohamed, T. Sami
Abstract:
Over the past few years, mobile communication technologies have seen significant evolution. This fact promoted the implementation of many systems in a multi-technological setting. From one system to another, the Quality of Service (QoS) provided to mobile consumers gets better. The growing number of normalized standards extends the available services for each consumer, moreover, most of the available radio frequencies have already been allocated, such as 3G, Wifi, Wimax, and LTE. A study by the Federal Communications Commission (FCC) found that certain frequency bands are partially occupied in particular locations and times. So, the idea of Cognitive Radio (CR) is to share the spectrum between a primary user (PU) and a secondary user (SU). The main objective of this spectrum management is to achieve a maximum rate of exploitation of the radio spectrum. In general, the CR can greatly improve the quality of service (QoS) and improve the reliability of the link. The problem will reside in the possibility of proposing a technique to improve the reliability of the wireless link by using the CR with some routing protocols. However, users declared that the links were unreliable and that it was an incompatibility with QoS. In our case, we choose the QoS parameter "bandwidth" to perform a supervised classification. In this paper, we propose a comparative study between some routing protocols, taking into account the variation of different technologies on the existing spectral bandwidth like 3G, WIFI, WIMAX, and LTE. Due to the simulation results, we observe that LTE has significantly higher availability bandwidth compared with other technologies. The performance of the OLSR protocol is better than other on-demand routing protocols (DSR, AODV and DSDV), in LTE technology because of the proper receiving of packets, less packet drop and the throughput. Numerous simulations of routing protocols have been made using simulators such as NS3.Keywords: cognitive radio, multi technology, network simulator (NS3), routing protocol
Procedia PDF Downloads 521279 Evaluation of Interaction Between Fans and Celebrities in New Media
Authors: Mohadese Motahari
Abstract:
In general, we consider the phenomenon of "fandism" or extreme fandom to be an aspect of fandom for a person, a group, or a collection, which leads to extreme support for them. So, for example, we consider a fan or a "fanatic" (which literally means a "fanatical person") to be a person who is extremely interested in a certain topic or topics and has a special passion and fascination for that issue. It may also be beyond the scope of logic and normal behavior of the society. With the expansion of the media and the advancement of technology, the phenomenon of fandom also underwent many changes and not only became more intense, but a large economy was also formed alongside it, and it is becoming more and more important every day. This economy, which emerged from the past with the formation of the first media, has now taken a different form with the development of media and social networks, as well as the change in the interaction between celebrities and audiences. Earning huge amounts of money with special methods in every social network and every media is achieved through fans and fandoms. In this article, we have studied the relationship between fans and famous people with reference to the economic debates surrounding it.Keywords: fandism, famous people, social media, new media
Procedia PDF Downloads 821278 Citation Analysis of New Zealand Court Decisions
Authors: Tobias Milz, L. Macpherson, Varvara Vetrova
Abstract:
The law is a fundamental pillar of human societies as it shapes, controls and governs how humans conduct business, behave and interact with each other. Recent advances in computer-assisted technologies such as NLP, data science and AI are creating opportunities to support the practice, research and study of this pervasive domain. It is therefore not surprising that there has been an increase in investments into supporting technologies for the legal industry (also known as “legal tech” or “law tech”) over the last decade. A sub-discipline of particular appeal is concerned with assisted legal research. Supporting law researchers and practitioners to retrieve information from the vast amount of ever-growing legal documentation is of natural interest to the legal research community. One tool that has been in use for this purpose since the early nineteenth century is legal citation indexing. Among other use cases, they provided an effective means to discover new precedent cases. Nowadays, computer-assisted network analysis tools can allow for new and more efficient ways to reveal the “hidden” information that is conveyed through citation behavior. Unfortunately, access to openly available legal data is still lacking in New Zealand and access to such networks is only commercially available via providers such as LexisNexis. Consequently, there is a need to create, analyze and provide a legal citation network with sufficient data to support legal research tasks. This paper describes the development and analysis of a legal citation Network for New Zealand containing over 300.000 decisions from 125 different courts of all areas of law and jurisdiction. Using python, the authors assembled web crawlers, scrapers and an OCR pipeline to collect and convert court decisions from openly available sources such as NZLII into uniform and machine-readable text. This facilitated the use of regular expressions to identify references to other court decisions from within the decision text. The data was then imported into a graph-based database (Neo4j) with the courts and their respective cases represented as nodes and the extracted citations as links. Furthermore, additional links between courts of connected cases were added to indicate an indirect citation between the courts. Neo4j, as a graph-based database, allows efficient querying and use of network algorithms such as PageRank to reveal the most influential/most cited courts and court decisions over time. This paper shows that the in-degree distribution of the New Zealand legal citation network resembles a power-law distribution, which indicates a possible scale-free behavior of the network. This is in line with findings of the respective citation networks of the U.S. Supreme Court, Austria and Germany. The authors of this paper provide the database as an openly available data source to support further legal research. The decision texts can be exported from the database to be used for NLP-related legal research, while the network can be used for in-depth analysis. For example, users of the database can specify the network algorithms and metrics to only include specific courts to filter the results to the area of law of interest.Keywords: case citation network, citation analysis, network analysis, Neo4j
Procedia PDF Downloads 971277 Cooperative Coevolution for Neuro-Evolution of Feed Forward Networks for Time Series Prediction Using Hidden Neuron Connections
Authors: Ravneil Nand
Abstract:
Cooperative coevolution uses problem decomposition methods to solve a larger problem. The problem decomposition deals with breaking down the larger problem into a number of smaller sub-problems depending on their method. Different problem decomposition methods have their own strengths and limitations depending on the neural network used and application problem. In this paper we are introducing a new problem decomposition method known as Hidden-Neuron Level Decomposition (HNL). The HNL method is competing with established problem decomposition method in time series prediction. The results show that the proposed approach has improved the results in some benchmark data sets when compared to the standalone method and has competitive results when compared to methods from literature.Keywords: cooperative coevaluation, feed forward network, problem decomposition, neuron, synapse
Procedia PDF Downloads 3211276 Deep Reinforcement Learning-Based Computation Offloading for 5G Vehicle-Aware Multi-Access Edge Computing Network
Authors: Ziying Wu, Danfeng Yan
Abstract:
Multi-Access Edge Computing (MEC) is one of the key technologies of the future 5G network. By deploying edge computing centers at the edge of wireless access network, the computation tasks can be offloaded to edge servers rather than the remote cloud server to meet the requirements of 5G low-latency and high-reliability application scenarios. Meanwhile, with the development of IOV (Internet of Vehicles) technology, various delay-sensitive and compute-intensive in-vehicle applications continue to appear. Compared with traditional internet business, these computation tasks have higher processing priority and lower delay requirements. In this paper, we design a 5G-based Vehicle-Aware Multi-Access Edge Computing Network (VAMECN) and propose a joint optimization problem of minimizing total system cost. In view of the problem, a deep reinforcement learning-based joint computation offloading and task migration optimization (JCOTM) algorithm is proposed, considering the influences of multiple factors such as concurrent multiple computation tasks, system computing resources distribution, and network communication bandwidth. And, the mixed integer nonlinear programming problem is described as a Markov Decision Process. Experiments show that our proposed algorithm can effectively reduce task processing delay and equipment energy consumption, optimize computing offloading and resource allocation schemes, and improve system resource utilization, compared with other computing offloading policies.Keywords: multi-access edge computing, computation offloading, 5th generation, vehicle-aware, deep reinforcement learning, deep q-network
Procedia PDF Downloads 991275 Information in Public Domain: How Far It Measures Government's Accountability
Authors: Sandip Mitra
Abstract:
Studies on Governance and Accountability has often stressed the need to release Data in public domain to increase transparency ,which otherwise act as an evidence of performance. However, inefficient handling, lack of capacity and the dynamics of transfers (especially fund transfers) are important issues which need appropriate attention. E-Governance alone can not serve as a measure of transparency as long as a comprehensive planning is instituted. Studies on Governance and public exposure has often triggered public opinion in favour or against any government. The root of the problem (especially in local governments) lies in the management of the governance. The participation of the people in the local government functioning, the networks within and outside the locality, synergy with various layers of Government are crucial in understanding the activities of any government. Unfortunately, data on such issues are not released in the public domain .If they are at all released , the extraction of information is often hindered for complicated designs. A Study has been undertaken with a few local Governments in India. The data has been analysed to substantiate the views.Keywords: accountability, e-governance, transparency, local government
Procedia PDF Downloads 4231274 Hybrid Hierarchical Clustering Approach for Community Detection in Social Network
Authors: Radhia Toujani, Jalel Akaichi
Abstract:
Social Networks generally present a hierarchy of communities. To determine these communities and the relationship between them, detection algorithms should be applied. Most of the existing algorithms, proposed for hierarchical communities identification, are based on either agglomerative clustering or divisive clustering. In this paper, we present a hybrid hierarchical clustering approach for community detection based on both bottom-up and bottom-down clustering. Obviously, our approach provides more relevant community structure than hierarchical method which considers only divisive or agglomerative clustering to identify communities. Moreover, we performed some comparative experiments to enhance the quality of the clustering results and to show the effectiveness of our algorithm.Keywords: agglomerative hierarchical clustering, community structure, divisive hierarchical clustering, hybrid hierarchical clustering, opinion mining, social network, social network analysis
Procedia PDF Downloads 3531273 Compensation of Power Quality Disturbances Using DVR
Authors: R. Rezaeipour
Abstract:
One of the key aspects of power quality improvement in power system is the mitigation of voltage sags/swells and flicker. Custom power devices have been known as the best tools for voltage disturbances mitigation as well as reactive power compensation. Dynamic voltage restorer (DVR) which is the most efficient and effective modern custom power device can provide the most commercial solution to solve several problems of power quality in distribution networks. This paper deals with analysis and simulation technique of DVR based on instantaneous power theory which is a quick control to detect signals. The main purpose of this work is to remove three important disturbances including voltage sags/swells and flicker. Simulation of the proposed method was carried out on two sample systems by using MATLAB software environment and the results of simulation show that the proposed method is able to provide desirable power quality in the presence of wide range of disturbances.Keywords: DVR, power quality, voltage sags, voltage swells, flicker
Procedia PDF Downloads 3361272 Iot-Based Interactive Patient Identification and Safety Management System
Authors: Jonghoon Chun, Insung Kim, Jonghyun Lim, Gun Ro
Abstract:
We believe that it is possible to provide a solution to reduce patient safety accidents by displaying correct medical records and prescription information through interactive patient identification. Our system is based on the use of smart bands worn by patients and these bands communicate with the hybrid gateways which understand both BLE and Wifi communication protocols. Through the convergence of low-power Bluetooth (BLE) and hybrid gateway technology, which is one of short-range wireless communication technologies, we implement ‘Intelligent Patient Identification and Location Tracking System’ to prevent medical malfunction frequently occurring in medical institutions. Based on big data and IOT technology using MongoDB, smart band (BLE, NFC function) and hybrid gateway, we develop a system to enable two-way communication between medical staff and hospitalized patients as well as to store locational information of the patients in minutes. Based on the precise information provided using big data systems, such as location tracking and movement of in-hospital patients wearing smart bands, our findings include the fact that a patient-specific location tracking algorithm can more efficiently operate HIS (Hospital Information System) and other related systems. Through the system, we can always correctly identify patients using identification tags. In addition, the system automatically determines whether the patient is a scheduled for medical service by the system in use at the medical institution, and displays the appropriateness of the medical treatment and the medical information (medical record and prescription information) on the screen and voice. This work was supported in part by the Korea Technology and Information Promotion Agency for SMEs (TIPA) grant funded by the Korean Small and Medium Business Administration (No. S2410390).Keywords: BLE, hybrid gateway, patient identification, IoT, safety management, smart band
Procedia PDF Downloads 3031271 Road Safety in the Great Britain: An Exploratory Data Analysis
Authors: Jatin Kumar Choudhary, Naren Rayala, Abbas Eslami Kiasari, Fahimeh Jafari
Abstract:
The Great Britain has one of the safest road networks in the world. However, the consequences of any death or serious injury are devastating for loved ones, as well as for those who help the severely injured. This paper aims to analyse the Great Britain's road safety situation and show the response measures for areas where the total damage caused by accidents can be significantly and quickly reduced. In this paper, we do an exploratory data analysis using STATS19 data. For the past 30 years, the UK has had a good record in reducing fatalities. The UK ranked third based on the number of road deaths per million inhabitants. There were around 165,000 accidents reported in the Great Britain in 2009 and it has been decreasing every year until 2019 which is under 120,000. The government continues to scale back road deaths empowering responsible road users by identifying and prosecuting the parameters that make the roads less safe.Keywords: road safety, data analysis, openstreetmap, feature expanding.
Procedia PDF Downloads 1281270 A Comparison of Image Data Representations for Local Stereo Matching
Authors: André Smith, Amr Abdel-Dayem
Abstract:
The stereo matching problem, while having been present for several decades, continues to be an active area of research. The goal of this research is to find correspondences between elements found in a set of stereoscopic images. With these pairings, it is possible to infer the distance of objects within a scene, relative to the observer. Advancements in this field have led to experimentations with various techniques, from graph-cut energy minimization to artificial neural networks. At the basis of these techniques is a cost function, which is used to evaluate the likelihood of a particular match between points in each image. While at its core, the cost is based on comparing the image pixel data; there is a general lack of consistency as to what image data representation to use. This paper presents an experimental analysis to compare the effectiveness of more common image data representations. The goal is to determine the effectiveness of these data representations to reduce the cost for the correct correspondence relative to other possible matches.Keywords: colour data, local stereo matching, stereo correspondence, disparity map
Procedia PDF Downloads 3611269 Measuring Organizational Resiliency for Flood Response in Thailand
Authors: Sudha Arlikatti, Laura Siebeneck, Simon A. Andrew
Abstract:
The objective of this research is to measure organizational resiliency through five attributes namely, rapidity, redundancy, resourcefulness, and robustness and to provide recommendations for resiliency building in flood risk communities. The research was conducted in Thailand following the severe floods of 2011 triggered by Tropical Storm Nock-ten. The floods lasted over eight months starting in June 2011 affecting 65 of the country’s 76 provinces and over 12 million people. Funding from a US National Science Foundation grant was used to collect ephemeral data in rural (Ayutthaya), suburban (Pathum Thani), and urban (Bangkok) provinces of Thailand. Semi-structured face-to-face interviews were conducted in Thai with 44 contacts from public, private, and non-profit organizations including universities, schools, automobile companies, vendors, tourist agencies, monks from temples, faith based organizations, and government agencies. Multiple triangulations were used to analyze the data by identifying selective themes from the qualitative data, validated with quantitative data and news media reports. This helped to obtain a more comprehensive view of how organizations in different geographic settings varied in their understanding of what enhanced or hindered their resilience and consequently their speed and capacities to respond. The findings suggest that the urban province of Bangkok scored highest in resourcefulness, rapidity of response, robustness, and ability to rebound. This is not surprising considering that it is the country’s capital and the seat of government, economic, military and tourism sectors. However, contrary to expectations all 44 respondents noted that the rural province of Ayutthaya was the fastest to recover amongst the three. Its organizations scored high on redundancy and rapidity of response due to the strength of social networks, a flood disaster sub-culture due to annual flooding, and the help provided by monks from and faith based organizations. Organizations in the suburban community of Pathum Thani scored lowest on rapidity of response and resourcefulness due to limited and ambiguous warnings, lack of prior flood experience and controversies that government flood protection works like sandbagging favored the capital city of Bangkok over them. Such a micro-level examination of organizational resilience in rural, suburban and urban areas in a country through mixed methods studies has its merits in getting a nuanced understanding of the importance of disaster subcultures and religious norms for resilience. This can help refocus attention on the strengths of social networks and social capital, for flood mitigation.Keywords: disaster subculture, flood response, organizational resilience, Thailand floods, religious beliefs and response, social capital and disasters
Procedia PDF Downloads 1501268 Decision-Making, Expectations and Life Project in Dependent Adults Due to Disability
Authors: Julia Córdoba
Abstract:
People are not completely autonomous, as we live in society; therefore, people could be defined as relationally dependent. The lack, decrease or loss of physical, psychological and/or social interdependence due to a disability situation is known as dependence. This is related to the need for help from another person in order to carry out activities of daily living. This population group lives with major social limitations that significantly reduce their participation and autonomy. They have high levels of stigma and invisibility from private environments (family and close networks), as well as from the public order (environment, community). The importance of this study lies in the fact that the lack of support and adjustments leads to what authors call the circle of exclusion. This circle describes how not accessing services - due to the difficulties caused by the disability situation impacts biological, social and psychological levels. This situation produces higher levels of exclusion and vulnerability. This study will focus on the process of autonomy and dependence of adults with disability from the model of disability proposed by the International Classification of Functioning, Health and Disability (ICF). The objectives are: i) to write down the relationship between autonomy and dependence based on socio-health variables and ii) to determine the relationship between the situation of autonomy and dependence and the expectations and interests of the participants. We propose a study that will use a survey technique through a previously validated virtual questionnaire. The data obtained will be analyzed using quantitative and qualitative methods for the details of the profiles obtained. No less than 200 questionnaires will be administered to people between 18 and 64 years of age who self-identify as having some degree of dependency due to disability. For the analysis of the results, the two main variables of autonomy and dependence will be considered. Socio-demographic variables such as age, gender identity, area of residence and family composition will be used. In relation to the biological dimension of the situation, the diagnosis, if any, and the type of disability will be asked. For the description of these profiles of autonomy and dependence, the following variables will be used: self-perception, decision-making, interests, expectations and life project, care of their health condition, support and social network, and labor and educational inclusion. The relationship between the target population and the variables collected provides several guidelines that could form the basis for the analysis of other research of interest in terms of self-perception, autonomy and dependence. The areas and situations where people state that they have greater possibilities to decide and have a say will be obtained. It will identify social (networks and support, educational background), demographic (age, gender identity and residence) and health-related variables (diagnosis and type of disability, quality of care) that may have a greater relationship with situations of dependency or autonomy. It will be studied whether the level of autonomy and/or dependence has an impact on the type of expectations and interests of the people surveyed.Keywords: life project, disability, inclusion, autonomy
Procedia PDF Downloads 601267 Identification of Breast Anomalies Based on Deep Convolutional Neural Networks and K-Nearest Neighbors
Authors: Ayyaz Hussain, Tariq Sadad
Abstract:
Breast cancer (BC) is one of the widespread ailments among females globally. The early prognosis of BC can decrease the mortality rate. Exact findings of benign tumors can avoid unnecessary biopsies and further treatments of patients under investigation. However, due to variations in images, it is a tough job to isolate cancerous cases from normal and benign ones. The machine learning technique is widely employed in the classification of BC pattern and prognosis. In this research, a deep convolution neural network (DCNN) called AlexNet architecture is employed to get more discriminative features from breast tissues. To achieve higher accuracy, K-nearest neighbor (KNN) classifiers are employed as a substitute for the softmax layer in deep learning. The proposed model is tested on a widely used breast image database called MIAS dataset for experimental purposes and achieved 99% accuracy.Keywords: breast cancer, DCNN, KNN, mammography
Procedia PDF Downloads 1261266 A Deep Reinforcement Learning-Based Secure Framework against Adversarial Attacks in Power System
Authors: Arshia Aflaki, Hadis Karimipour, Anik Islam
Abstract:
Generative Adversarial Attacks (GAAs) threaten critical sectors, ranging from fingerprint recognition to industrial control systems. Existing Deep Learning (DL) algorithms are not robust enough against this kind of cyber-attack. As one of the most critical industries in the world, the power grid is not an exception. In this study, a Deep Reinforcement Learning-based (DRL) framework assisting the DL model to improve the robustness of the model against generative adversarial attacks is proposed. Real-world smart grid stability data, as an IIoT dataset, test our method and improves the classification accuracy of a deep learning model from around 57 percent to 96 percent.Keywords: generative adversarial attack, deep reinforcement learning, deep learning, IIoT, generative adversarial networks, power system
Procedia PDF Downloads 171265 Flexible and Color Tunable Inorganic Light Emitting Diode Array for High Resolution Optogenetic Devices
Authors: Keundong Lee, Dongha Yoo, Youngbin Tchoe, Gyu-Chul Yi
Abstract:
Light emitting diode (LED) array is an ideal optical stimulation tool for optogenetics, which controls inhibition and excitation of specific neurons with light-sensitive ion channels or pumps. Although a fiber-optic cable with an external light source, either a laser or LED mechanically connected to the end of the fiber-optic cable has widely been used for illumination on neural tissue, a new approach to use micro LEDs (µLEDs) has recently been demonstrated. The LEDs can be placed directly either on the cortical surface or within the deep brain using a penetrating depth probe. Accordingly, this method would not need a permanent opening in the skull if the LEDs are integrated with miniature electrical power source and wireless communication. In addition, multiple color generation from single µLED cell would enable to excite and/or inhibit neurons in localized regions. Here, we demonstrate flexible and color tunable µLEDs for the optogenetic device applications. The flexible and color tunable LEDs was fabricated using multifaceted gallium nitride (GaN) nanorod arrays with GaN nanorods grown on InxGa1−xN/GaN single quantum well structures (SQW) anisotropically formed on the nanorod tips and sidewalls. For various electroluminescence (EL) colors, current injection paths were controlled through a continuous p-GaN layer depending on the applied bias voltage. The electric current was injected through different thickness and composition, thus changing the color of light from red to blue that the LED emits. We believe that the flexible and color tunable µLEDs enable us to control activities of the neuron by emitting various colors from the single µLED cell.Keywords: light emitting diode, optogenetics, graphene, flexible optoelectronics
Procedia PDF Downloads 2061264 Ontologies for Social Media Digital Evidence
Authors: Edlira Kalemi, Sule Yildirim-Yayilgan
Abstract:
Online Social Networks (OSNs) are nowadays being used widely and intensively for crime investigation and prevention activities. As they provide a lot of information they are used by the law enforcement and intelligence. An extensive review on existing solutions and models for collecting intelligence from this source of information and making use of it for solving crimes has been presented in this article. The main focus is on smart solutions and models where ontologies have been used as the main approach for representing criminal domain knowledge. A framework for a prototype ontology named SC-Ont will be described. This defines terms of the criminal domain ontology and the relations between them. The terms and the relations are extracted during both this review and the discussions carried out with domain experts. The development of SC-Ont is still ongoing work, where in this paper, we report mainly on the motivation for using smart ontology models and the possible benefits of using them for solving crimes.Keywords: criminal digital evidence, social media, ontologies, reasoning
Procedia PDF Downloads 3811263 Characterization of Internet Exchange Points by Using Quantitative Data
Authors: Yamba Dabone, Tounwendyam Frédéric Ouedraogo, Pengwendé Justin Kouraogo, Oumarou Sie
Abstract:
Reliable data transport over the Internet is one of the goals of researchers in the field of computer science. Data such as videos and audio files are becoming increasingly large. As a result, transporting them over the Internet is becoming difficult. Therefore, it has been important to establish a method to locally interconnect autonomous systems (AS) with each other to facilitate traffic exchange. It is in this context that Internet Exchange Points (IXPs) are set up to facilitate local and even regional traffic. They are now the lifeblood of the Internet. Therefore, it is important to think about the factors that can characterize IXPs. However, other more quantifiable characteristics can help determine the quality of an IXP. In addition, these characteristics may allow ISPs to have a clearer view of the exchange node and may also convince other networks to connect to an IXP. To that end, we define five new IXP characteristics: the attraction rate (τₐₜₜᵣ); and the peering rate (τₚₑₑᵣ); the target rate of an IXP (Objₐₜₜ); the number of IXP links (Nₗᵢₙₖ); the resistance rate τₑ𝒻𝒻 and the attraction failure rate (τ𝒻).Keywords: characteristic, autonomous system, internet service provider, internet exchange point, rate
Procedia PDF Downloads 861262 A Reasoning Method of Cyber-Attack Attribution Based on Threat Intelligence
Authors: Li Qiang, Yang Ze-Ming, Liu Bao-Xu, Jiang Zheng-Wei
Abstract:
With the increasing complexity of cyberspace security, the cyber-attack attribution has become an important challenge of the security protection systems. The difficult points of cyber-attack attribution were forced on the problems of huge data handling and key data missing. According to this situation, this paper presented a reasoning method of cyber-attack attribution based on threat intelligence. The method utilizes the intrusion kill chain model and Bayesian network to build attack chain and evidence chain of cyber-attack on threat intelligence platform through data calculation, analysis and reasoning. Then, we used a number of cyber-attack events which we have observed and analyzed to test the reasoning method and demo system, the result of testing indicates that the reasoning method can provide certain help in cyber-attack attribution.Keywords: reasoning, Bayesian networks, cyber-attack attribution, Kill Chain, threat intelligence
Procedia PDF Downloads 4411261 Cybervetting and Online Privacy in Job Recruitment – Perspectives on the Current and Future Legislative Framework Within the EU
Authors: Nicole Christiansen, Hanne Marie Motzfeldt
Abstract:
In recent years, more and more HR professionals have been using cyber-vetting in job recruitment in an effort to find the perfect match for the company. These practices are growing rapidly, accessing a vast amount of data from social networks, some of which is privileged and protected information. Thus, there is a risk that the right to privacy is becoming a duty to manage your private data. This paper investigates to which degree a job applicant's fundamental rights are protected adequately in current and future legislation in the EU. This paper argues that current data protection regulations and forthcoming regulations on the use of AI ensure sufficient protection. However, even though the regulation on paper protects employees within the EU, the recruitment sector may not pay sufficient attention to the regulation as it not specifically targeting this area. Therefore, the lack of specific labor and employment regulation is a concern that the social partners should attend to.Keywords: AI, cyber vetting, data protection, job recruitment, online privacy
Procedia PDF Downloads 781260 Assessment of Taiwan Railway Occurrences Investigations Using Causal Factor Analysis System and Bayesian Network Modeling Method
Authors: Lee Yan Nian
Abstract:
Safety investigation is different from an administrative investigation in that the former is conducted by an independent agency and the purpose of such investigation is to prevent accidents in the future and not to apportion blame or determine liability. Before October 2018, Taiwan railway occurrences were investigated by local supervisory authority. Characteristics of this kind of investigation are that enforcement actions, such as administrative penalty, are usually imposed on those persons or units involved in occurrence. On October 21, 2018, due to a Taiwan Railway accident, which caused 18 fatalities and injured another 267, establishing an agency to independently investigate this catastrophic railway accident was quickly decided. The Taiwan Transportation Safety Board (TTSB) was then established on August 1, 2019 to take charge of investigating major aviation, marine, railway and highway occurrences. The objective of this study is to assess the effectiveness of safety investigations conducted by the TTSB. In this study, the major railway occurrence investigation reports published by the TTSB are used for modeling and analysis. According to the classification of railway occurrences investigated by the TTSB, accident types of Taiwan railway occurrences can be categorized into: derailment, fire, Signal Passed at Danger and others. A Causal Factor Analysis System (CFAS) developed by the TTSB is used to identify the influencing causal factors and their causal relationships in the investigation reports. All terminologies used in the CFAS are equivalent to the Human Factors Analysis and Classification System (HFACS) terminologies, except for “Technical Events” which was added to classify causal factors resulting from mechanical failure. Accordingly, the Bayesian network structure of each occurrence category is established based on the identified causal factors in the CFAS. In the Bayesian networks, the prior probabilities of identified causal factors are obtained from the number of times in the investigation reports. Conditional Probability Table of each parent node is determined from domain experts’ experience and judgement. The resulting networks are quantitatively assessed under different scenarios to evaluate their forward predictions and backward diagnostic capabilities. Finally, the established Bayesian network of derailment is assessed using investigation reports of the same accident which was investigated by the TTSB and the local supervisory authority respectively. Based on the assessment results, findings of the administrative investigation is more closely tied to errors of front line personnel than to organizational related factors. Safety investigation can identify not only unsafe acts of individual but also in-depth causal factors of organizational influences. The results show that the proposed methodology can identify differences between safety investigation and administrative investigation. Therefore, effective intervention strategies in associated areas can be better addressed for safety improvement and future accident prevention through safety investigation.Keywords: administrative investigation, bayesian network, causal factor analysis system, safety investigation
Procedia PDF Downloads 1071259 Evaluation of Modern Natural Language Processing Techniques via Measuring a Company's Public Perception
Authors: Burak Oksuzoglu, Savas Yildirim, Ferhat Kutlu
Abstract:
Opinion mining (OM) is one of the natural language processing (NLP) problems to determine the polarity of opinions, mostly represented on a positive-neutral-negative axis. The data for OM is usually collected from various social media platforms. In an era where social media has considerable control over companies’ futures, it’s worth understanding social media and taking actions accordingly. OM comes to the fore here as the scale of the discussion about companies increases, and it becomes unfeasible to gauge opinion on individual levels. Thus, the companies opt to automize this process by applying machine learning (ML) approaches to their data. For the last two decades, OM or sentiment analysis (SA) has been mainly performed by applying ML classification algorithms such as support vector machines (SVM) and Naïve Bayes to a bag of n-gram representations of textual data. With the advent of deep learning and its apparent success in NLP, traditional methods have become obsolete. Transfer learning paradigm that has been commonly used in computer vision (CV) problems started to shape NLP approaches and language models (LM) lately. This gave a sudden rise to the usage of the pretrained language model (PTM), which contains language representations that are obtained by training it on the large datasets using self-supervised learning objectives. The PTMs are further fine-tuned by a specialized downstream task dataset to produce efficient models for various NLP tasks such as OM, NER (Named-Entity Recognition), Question Answering (QA), and so forth. In this study, the traditional and modern NLP approaches have been evaluated for OM by using a sizable corpus belonging to a large private company containing about 76,000 comments in Turkish: SVM with a bag of n-grams, and two chosen pre-trained models, multilingual universal sentence encoder (MUSE) and bidirectional encoder representations from transformers (BERT). The MUSE model is a multilingual model that supports 16 languages, including Turkish, and it is based on convolutional neural networks. The BERT is a monolingual model in our case and transformers-based neural networks. It uses a masked language model and next sentence prediction tasks that allow the bidirectional training of the transformers. During the training phase of the architecture, pre-processing operations such as morphological parsing, stemming, and spelling correction was not used since the experiments showed that their contribution to the model performance was found insignificant even though Turkish is a highly agglutinative and inflective language. The results show that usage of deep learning methods with pre-trained models and fine-tuning achieve about 11% improvement over SVM for OM. The BERT model achieved around 94% prediction accuracy while the MUSE model achieved around 88% and SVM did around 83%. The MUSE multilingual model shows better results than SVM, but it still performs worse than the monolingual BERT model.Keywords: BERT, MUSE, opinion mining, pretrained language model, SVM, Turkish
Procedia PDF Downloads 1301258 Radar Signal Detection Using Neural Networks in Log-Normal Clutter for Multiple Targets Situations
Authors: Boudemagh Naime
Abstract:
Automatic radar detection requires some methods of adapting to variations in the background clutter in order to control their false alarm rate. The problem becomes more complicated in non-Gaussian environment. In fact, the conventional approach in real time applications requires a complex statistical modeling and much computational operations. To overcome these constraints, we propose another approach based on artificial neural network (ANN-CMLD-CFAR) using a Back Propagation (BP) training algorithm. The considered environment follows a log-normal distribution in the presence of multiple Rayleigh-targets. To evaluate the performances of the considered detector, several situations, such as scale parameter and the number of interferes targets, have been investigated. The simulation results show that the ANN-CMLD-CFAR processor outperforms the conventional statistical one.Keywords: radat detection, ANN-CMLD-CFAR, log-normal clutter, statistical modelling
Procedia PDF Downloads 3531257 Rumination Time and Reticuloruminal Temperature around Calving in Eutocic and Dystocic Dairy Cows
Authors: Levente Kovács, Fruzsina Luca Kézér, Ottó Szenci
Abstract:
Prediction of the onset of calving and recognizing difficulties at calving has great importance in decreasing neonatal losses and reducing the risk of health problems in the early postpartum period. In this study, changes of rumination time, reticuloruminal pH and temperature were investigated in eutocic (EUT, n = 10) and dystocic (DYS, n = 8) dairy cows around parturition. Rumination time was continuously recorded using an acoustic biotelemetry system, whereas reticuloruminal pH and temperature were recorded using an indwelling and wireless data transmitting system. The recording period lasted from 3 d before calving until 7 days in milk. For the comparison of rumination time and reticuloruminal characteristics between groups, time to return to baseline (the time interval required to return to baseline from the delivery of the calf) and area under the curve (AUC, both for prepartum and postpartum periods) were calculated for each parameter. Rumination time decreased from baseline 28 h before calving both for EUT and DYS cows (P = 0.023 and P = 0.017, respectively). After 20 h before calving, it decreased onwards to reach 32.4 ± 2.3 and 13.2 ± 2.0 min/4 h between 8 and 4 h before delivery in EUT and DYS cows, respectively, and then it decreased below 10 and 5 min during the last 4 h before calving (P = 0.003 and P = 0.008, respectively). Until 12 h after delivery rumination time reached 42.6 ± 2.7 and 51.0 ± 3.1 min/4 h in DYS and EUT dams, respectively, however, AUC and time to return to baseline suggested lower rumination activity in DYS cows than in EUT dams for the 168-h postpartum observational period (P = 0.012 and P = 0.002, respectively). Reticuloruminal pH decreased from baseline 56 h before calving both for EUT and DYS cows (P = 0.012 and P = 0.016, respectively), but did not differ between groups before delivery. In DYS cows, reticuloruminal temperature decreased from baseline 32 h before calving by 0.23 ± 0.02 °C (P = 0.012), whereas in EUT cows such a decrease was found only 20 h before delivery (0.48 ± 0.05 °C, P < 0.01). AUC of reticuloruminal temperature calculated for the prepartum period was greater in EUT cows than in DYS cows (P = 0.042). During the first 4 h after calving, it decreased from 39.7 ± 0.1 to 39.00 ± 0.1 °C and from 39.8 ± 0.1 to 38.8 ± 0.1 °C in EUT and DYS cows, respectively (P < 0.01 for both groups) and reached baseline levels after 35.4 ± 3.4 and 37.8 ± 4.2 h after calving in EUT and DYS cows, respectively. Based on our results, continuous monitoring of changes in rumination time and reticuloruminal temperature seems to be promising in the early detection of cows with a higher risk of dystocia. Depressed postpartum rumination time of DYS cows highlights the importance of the monitoring of cows experiencing difficulties at calving.Keywords: reticuloruminal pH, reticuloruminal temperature, rumination time, dairy cows, dystocia
Procedia PDF Downloads 308