Search results for: restricted boltzmann machine

Authors: Jaroslaw Mizera, Pawel Wisniewski, Ryszard Sitek

Abstract:

Selective Laser Melting (SLM) is an advanced additive manufacturing technique used for producing parts made of wide range of materials such as: austenitic steel, titanium, nickel etc. In the our experiment we produced a Ti-AlN composite from a mixture of titanium and aluminum nitride respectively 70% at. and 30% at. using SLM technique. In order to define the size of powder particles, laser diffraction tests were performed on HORIBA LA-950 device. The microstructure and chemical composition of the composite was examined by Scanning Electron Microscopy (SEM). The chemical composition in micro areas of the obtained samples was determined by of EDS. The phase composition was analyzed by X-ray phase analysis (XRD). Microhardness Vickers tests were performed using Zwick/Roell microhardness machine under the load of 0.2kG (HV0.2). Hardness measurements were made along the building (xy) and along the plane of the lateral side of the cuboid (xz). The powder used for manufacturing of the samples had a mean particle size of 41μm. It was homogenous with a spherical shape. The specimens were built chiefly from Ti, TiN and AlN. The dendritic microstructure was porous and fine-grained. Some of the aluminum nitride remained unmelted but no porosity was observed in the interface. The formed material was characterized by high hardness exceeding 700 HV0.2 over the entire cross-section.

Keywords: Selective Laser Melting, Composite, SEM, microhardness

Procedia PDF Downloads 135

Authors: Aakansha Gupta, Neha Vadnere, Tapasvi Soni, M. Anbarsi

Abstract:

Protein secondary structure prediction is one of the most important goals pursued by bioinformatics and theoretical chemistry; it is highly important in medicine and biotechnology. Some aspects of protein functions and genome analysis can be predicted by secondary structure prediction. This is used to help annotate sequences, classify proteins, identify domains, and recognize functional motifs. In this paper, we represent protein secondary structure as a mathematical model. To extract and predict the protein secondary structure from the primary structure, we require a set of parameters. Any constants appearing in the model are specified by these parameters, which also provide a mechanism for efficient and accurate use of data. To estimate these model parameters there are many algorithms out of which the most popular one is the EM algorithm or called the Expectation Maximization Algorithm. These model parameters are estimated with the use of protein datasets like RS126 by using the Bayesian Probabilistic method (data set being categorical). This paper can then be extended into comparing the efficiency of EM algorithm to the other algorithms for estimating the model parameters, which will in turn lead to an efficient component for the Protein Secondary Structure Prediction. Further this paper provides a scope to use these parameters for predicting secondary structure of proteins using machine learning techniques like neural networks and fuzzy logic. The ultimate objective will be to obtain greater accuracy better than the previously achieved.

Keywords: model parameters, expectation maximization algorithm, protein secondary structure prediction, bioinformatics

Procedia PDF Downloads 479

Authors: Adriana de Paula Lacerda Santos, Bruna Godke, Mauro Lacerda Santos Filho

Abstract:

Several dams in the world have already collapsed, causing environmental, social and economic damage. The concern to avoid future disasters has stimulated the creation of a great number of laws and rules in many countries. In Brazil, Law 12.334/2010 was created, which establishes the National Policy on Dam Safety. Overall, this policy requires the dam owners to invest in the maintenance of their structures and to improve its monitoring systems in order to provide faster and straightforward responses in the case of an increase of risks. As monitoring tools, visual inspections has provides comprehensive assessment of the structures performance, while auscultation’s instrumentation has added specific information on operational or behavioral changes, providing an alarm when a performance indicator exceeds the acceptable limits. These limits can be set using statistical methods based on the relationship between instruments measures and other variables, such as reservoir level, time of the year or others instruments measuring. Besides the design parameters (uplift of the foundation, displacements, etc.) the dam instrumentation can also be used to monitor the behavior of defects and damage manifestations. Specifically in concrete gravity dams, one of the main causes for the appearance of cracks, are the concrete volumetric changes generated by the thermal origin phenomena, which are associated with the construction process of these structures. Based on this, the goal of this research is to propose a monitoring process of the thermal cracking behavior in concrete gravity dams, through the instrumentation data analysis and the establishment of control values. Therefore, as a case study was selected the Block B-11 of José Richa Governor Dam Power Plant, that presents a cracking process, which was identified even before filling the reservoir in August’ 1998, and where crack meters and surface thermometers were installed for its monitoring. Although these instruments were installed in May 2004, the research was restricted to study the last 4.5 years (June 2010 to November 2014), when all the instruments were calibrated and producing reliable data. The adopted method is based on simple linear correlations procedures to understand the interactions among the instruments time series, verifying the response times between them. The scatter plots were drafted from the best correlations, which supported the definition of the limit control values. Among the conclusions, it is shown that there is a strong or very strong correlation between ambient temperature and the crack meters and flowmeters measurements. Based on the results of the statistical analysis, it was possible to develop a tool for monitoring the behavior of the case study cracks. Thus it was fulfilled the goal of the research to develop a proposal for a monitoring process of the behavior of thermal cracking in concrete gravity dams.

Keywords: concrete gravity dam, dams safety, instrumentation, simple linear correlation

Procedia PDF Downloads 290

Authors: Ramadas Narayanan, Surya Bhattrai, Vu Hoan

Abstract:

Threshing and cleaning are crucial post-harvest procedures that are carried out to separate the grain or seed from the harvested plant and eliminate any potential contaminants or foreign debris. After harvesting, threshing and cleaning are necessary for the clean seeds to guarantee high quality and acceptable for consumption or further processing. For mechanised production, threshing can be conducted in a thresher. Afterwards, the seeds are to be cleaned in dedicated seed-cleaning facilities. This research investigates the effectiveness of Kimseed cleaning equipment MK3, designed for grain crops for processing new crops such as sesame, fennel and kalonji. Subsequently, systematic trials were conducted to adapt the equipment to the applications in sesame and spice crops. It was done to develop methods for mechanising harvest and post-harvest operations. For sesame, it is recommended to have t a two-step process in the cleaning machine to remove large and small contaminants. The first step is to remove the large contaminants, and the second is to remove the smaller ones. The optimal parameters for cleaning fennel are a shaker frequency of 6.0 to 6.5 Hz and an airflow of 1.0 to 1.5 m/s. The optimal parameters for cleaning kalonji are a shaker frequency of 5.5Hz to 6.0 Hz and airflow of 1.0 to under 1.5m/s.

Keywords: sustainable mechanisation, sead cleaning process, optimal setting, shaker frequency

Procedia PDF Downloads 71

Authors: Seong Jun Yoon, Jasurbek Doliev, Sang Min Oh, Rodi Hartono, Kyoojae Shin

Abstract:

Electric power steering (EPS), which is commonly used in electric vehicles recently, is an electric-driven steering device for vehicles. Compared to hydraulic systems, EPS offers advantages such as simple system components, easy maintenance, and improved steering performance. However, because the EPS system is a nonlinear model, difficult problems arise in controller design. To address these, various machine learning and artificial intelligence approaches, notably artificial neural networks (ANN), have been applied. ANN can effectively determine relationships between inputs and outputs in a data-driven manner. This research explores two main areas: designing an EPS identifier using an ANN-based backpropagation (BP) algorithm and enhancing the EPS system controller with an ANN-based Levenberg-Marquardt (LM) algorithm. The proposed ANN-based BP algorithm shows superior performance and accuracy compared to linear transfer function estimators, while the LM algorithm offers better input angle reference tracking and faster response times than traditional PID controllers. Overall, the proposed ANN methods demonstrate significant promise in improving EPS system performance.

Keywords: ANN backpropagation modelling, electric power steering, transfer function estimator, electrical vehicle driving system

Procedia PDF Downloads 41

Authors: Ioannis Andrianakis, Panagiotis Panagiotopoulos, Kyriakos Chatzidimitriou, Dimitrios Tampakis, Manolis Falelakis

Abstract:

Live music, whether performed in organized venues, restaurants, hotels or any other spots, creates value chains that support and develop local economies and tourism development. In this paper, we describe PerformAnts, a platform that increases the mobility of musicians and their accessibility to remotely located venues by rationalizing the cost of live acts. By analyzing the event history and taking into account their potential availability, the platform provides bespoke recommendations to both bands and venues while also facilitating the organization of tours and helping rationalize transportation expenses by realizing an innovative mechanism called “chain booking”. Moreover, the platform provides an environment where complicated tasks such as technical and financial negotiations, concert promotion or copyrights are easily manipulated by users using best practices. The proposed solution provides important benefits to the whole spectrum of small/medium size concert organizers, as the complexity and the cost of the production are rationalized. The environment is also very beneficial for local talent, musicians that are very mobile, venues located away from large urban areas or in touristic destinations, and managers who will be in a position to coordinate a larger number of musicians without extra effort.

Keywords: machine learning, music industry, creative industries, web applications

Procedia PDF Downloads 94

Authors: Andres M. Tuates Jr., Ofero A. Caparino

Abstract:

Plastics have achieved a dominant position in agriculture because of their transparency, lightness in weight, impermeability to water and their resistance to microbial attack. However, this generates a higher quantity of wastes that are difficult to dispose of by farmers. To address these problems, the project aim to develop and evaluate the biodegradable film for mango fruit bag during development. The PBS and starch were melt-blended in a twin-screw extruder and then blown into film extrusion machine. The physic-chemical-mechanical properties of biodegradable fruit bag were done following standard methods of test. Field testing of fruit bag was also conducted to evaluate its durability and efficiency field condition. The PHilMech-FiC fruit bag is made of biodegradable material measuring 6 x 8 inches with a thickness of 150 microns. The tensile strength is within the range of LDPE while the elongation is within the range of HDPE. It is projected that after thirty-six (36) weeks, the film will be totally degraded. Results of field testing show that the quality of harvested fruits using PHilMech-FiC biodegradable fruit bag in terms of percent marketable, non-marketable and export, peel color at the ripe stage, flesh color, TSS, oBrix, percent edible portion is comparable with the existing bagging materials such as Chinese brown paper bag and old newspaper.

Keywords: cassava starch, PBS, biodegradable, chemical, mechanical properties

Procedia PDF Downloads 276

Authors: Rizwan Rizwan

Abstract:

This paper provides an overview of Federated Learning (FL) and its application in enhancing data security, privacy, and efficiency. FL utilizes three distinct architectures to ensure privacy is never compromised. It involves training individual edge devices and aggregating their models on a server without sharing raw data. This approach not only provides secure models without data sharing but also offers a highly efficient privacy--preserving solution with improved security and data access. Also we discusses various frameworks used in FL and its integration with machine learning, deep learning, and data mining. In order to address the challenges of multi--party collaborative modeling scenarios, a brief review FL scheme combined with an adaptive gradient descent strategy and differential privacy mechanism. The adaptive learning rate algorithm adjusts the gradient descent process to avoid issues such as model overfitting and fluctuations, thereby enhancing modeling efficiency and performance in multi-party computation scenarios. Additionally, to cater to ultra-large-scale distributed secure computing, the research introduces a differential privacy mechanism that defends against various background knowledge attacks.

Keywords: federated learning, differential privacy, gradient descent strategy, convergence, stability, threats

Procedia PDF Downloads 29

Authors: Nan Xu

Abstract:

In airlines, the crew scheduling problem is usually decomposed into two stages: crew pairing and crew rostering. In the crew pairing stage, pairings are generated such that each flight is covered by exactly one pairing and the overall cost is minimized. In the crew rostering stage, the pairings generated in the crew pairing stage are combined with off days, training and other breaks to create individual work schedules. The paper focuses on cabin crew rostering problem, which is challenging due to the extremely large size and the complex working rules involved. In our approach, the objective of rostering consists of two major components. The first is to minimize the number of unassigned pairings and the second is to ensure the fairness to crew members. There are two measures of fairness to crew members, the number of overnight duties and the total fly-hour over a given period. Pairings should be assigned to each crew member so that their actual overnight duties and fly hours are as close to the expected average as possible. Deviations from the expected average are penalized in the objective function. Since several small deviations are preferred than a large deviation, the penalization is quadratic. Our model of the airline crew rostering problem is based on column generation. The problem is decomposed into a master problem and subproblems. The mater problem is modeled as a set partition problem and exactly one roster for each crew is picked up such that the pairings are covered. The restricted linear master problem (RLMP) is considered. The current subproblem tries to find columns with negative reduced costs and add them to the RLMP for the next iteration. When no column with negative reduced cost can be found or a stop criteria is met, the procedure ends. The subproblem is to generate feasible crew rosters for each crew member. A separate acyclic weighted graph is constructed for each crew member and the subproblem is modeled as resource constrained shortest path problems in the graph. Labeling algorithm is used to solve it. Since the penalization is quadratic, a method to deal with non-additive shortest path problem using labeling algorithm is proposed and corresponding domination condition is defined. The major contribution of our model is: 1) We propose a method to deal with non-additive shortest path problem; 2) Operation to allow relaxing some soft rules is allowed in our algorithm, which can improve the coverage rate; 3) Multi-thread techniques are used to improve the efficiency of the algorithm when generating Line-of-Work for crew members. Here a column generation based algorithm for the airline cabin crew rostering problem is proposed. The objective is to assign a personalized roster to crew member which minimize the number of unassigned pairings and ensure the fairness to crew members. The algorithm we propose in this paper has been put into production in a major airline in China and numerical experiments show that it has a good performance.

Keywords: aircrew rostering, aircrew scheduling, column generation, SPPRC

Procedia PDF Downloads 145

Authors: Junkyeong Kim, Jooyoung Park, Aoqi Zhang, Seunghee Park

Abstract:

The tensile force of Pre-Stressed Concrete (PSC) girder is the most important factor for evaluating the performance of PSC girder bridges. To measure the tensile force of PSC girder, several NDT methods were studied. However, conventional NDT method cannot be applied to the real-size PSC girder because the PS tendons could not be approached. To measure the tensile force of real-size PSC girder, this study proposed embedded EM sensor based tensile force estimation method. The embedded EM sensor could be installed inside of PSC girder as a sheath joint before the concrete casting. After curing process, the PS tendons were installed, and the tensile force was induced step by step using hydraulic jacking machine. The B-H loop was measured using embedded EM sensor at each tensile force steps and to compare with actual tensile force, the load cell was installed at each end of girder. The magnetization energy loss, that is the closed area of B-H loop, was decreased according to the increase of tensile force with regular pattern. Thus, the tensile force could be estimated by the tracking the change of magnetization energy loss of PS tendons. Through the experimental result, the proposed method can be used to estimate the tensile force of the in-situ real-size PSC girder bridge.

Keywords: tensile force estimation, embedded EM sensor, magnetization energy loss, PSC girder

Procedia PDF Downloads 335

Authors: Felix Boehnisch, Alexander Lutz

Abstract:

Based on the number of users and the amount of content posted daily, Facebook is considered the largest social network in the world. This content includes images or text posts from companies but also private persons, which are also commented on by other users. However, it can sometimes be difficult for companies to keep track of all the posts and the reactions to them, especially when there are several posts a day that contain hundreds to thousands of comments. To facilitate this, the following paper deals with the possible applications of sentiment analysis to social media comments in order to be able to support the work in social media marketing. In a first step, post comments were divided into positive and negative by a subjective rating, then the same comments were checked for their polarity value by the two german python libraries TextBlobDE and SentiWS and also grouped into positive, negative, or even neutral. As a control, the subjective classifications were compared with the machine-generated ones by a confusion matrix, and relevant quality criteria were determined. The accuracy of both libraries was not really meaningful, with 60% to 66%. However, many words or sentences were not evaluated at all, so there seems to be room for optimization to possibly get more accurate results. In future studies, the use of these specific German libraries can be optimized to gain better insights by either applying them to stricter cleaned data or by adding a sentiment value to emojis, which have been removed from the comments in advance, as they are not contained in the libraries.

Keywords: Facebook, German libraries, polarity, sentiment analysis, social media comments

Procedia PDF Downloads 181

Authors: James G. Matthew, Seonmin Huh, Frank X. Bennett

Abstract:

English for Specific Purposes, ESP, aims to equip learners with necessary English language skills. Many ESP programs address language skills for job performance, including reading job related documents and oral proficiency. Within ESP is English for occupational purposes, EOP, which centers around developing communicative competence for the globalized workplace. Many ESP and EOP courses lack the content needed to assist students to progress at work, resulting in the need to create lexical compilation for different professions. It is important to teach communicative competence and soft skills for real job-related problem situations and address the complexities of the real world to help students to be successful in their professions. ESP and EOP research is therefore trying to balance both profession-specific educational contents as well as international multi-disciplinary language skills for the globalized workforce. The current study will build upon the existing discussion by developing pedagogy to assist students in their career through developing a strong practical command of relevant English vocabulary. Our research question focuses on the pedagogy two professors incorporated in their English for employment courses. The current study is a qualitative case study on the modes of teaching delivery for EOP in South Korea. Two foreign professors teaching at two different universities in South Korea volunteered for the study to explore their teaching practices. Both professors’ curriculums included the components of employment-related concept vocabulary, business presentations, CV/resume and cover letter preparation, and job interview preparation. All the pre-made recorded video lectures, live online class sessions with students, teachers’ lesson plans, teachers’ class materials, students’ assignments, and midterm and finals video conferences were collected for data analysis. The study then focused on unpacking representative patterns in their teaching methods. The professors used their strengths as native speakers to extend the class discussion from narrow and restricted conversations to giving students broader opportunities to practice authentic English conversation. The methods of teaching utilized three main steps to extend the conversation. Firstly, students were taught concept vocabulary. Secondly, the vocabulary was then combined in speaking activities where students had to solve scenarios, and the students were required to expand on the given forms of words and language expressions. Lastly, the students had conversations in English, using the language learnt. The conversations observed in both classes were those of authentic, expanded English communication and this way of expanding concept vocabulary lessons into extended conversation is one representative pedagogical approach that both professors took. Extended English conversation, therefore, is crucial for EOP education.

Keywords: concept vocabulary, english as a foreign language, english for employment, extended conversation

Procedia PDF Downloads 92

Authors: Hamidreza Ghazvinian, Khosro Ghazvinian, Touba Khodaiean

Abstract:

The importance of evaporation estimation in water resources and agricultural studies is undeniable. Pan evaporation are used as an indicator to determine the evaporation of lakes and reservoirs around the world due to the ease of interpreting its data. In this research, intelligent models were investigated in estimating pan evaporation on a daily basis. Shahroud and Mayamey were considered as the studied cities. These two cities are located in Semnan province in Iran. The mentioned cities have dry weather conditions that are susceptible to high evaporation potential. Meteorological data of 11 years of synoptic stations of Shahrood and Mayamey cities were used. The intelligent models used in this study are Artificial Neural Network (ANN), Least Squares Support Vector Machine (LSSVM), and M5 tree models. Meteorological parameters of minimum and maximum air temperature (Tmax, Tmin), wind speed (WS), sunshine hours (SH), air pressure (PA), relative humidity (RH) as selected input data and evaporation data from pan (EP) to The output data was considered. 70% of data is used at the education level, and 30 % of the data is used at the test level. Models used with explanation coefficient evaluation (R2) Root of Mean Squares Error (RMSE) and Mean Absolute Error (MAE). The results for the two Shahroud and Mayamey stations showed that the above three models' operations are rather appropriate.

Keywords: pan evaporation, intelligent methods, shahroud, mayamey

Procedia PDF Downloads 73

Authors: Amin Nezarat, Naeime Seifadini

Abstract:

Trip mode selection is a behavioral characteristic of passengers with immense importance for travel demand analysis, transportation planning, and traffic management. Identification of trip mode distribution will allow transportation authorities to adopt appropriate strategies to reduce travel time, traffic and air pollution. The majority of existing trip mode inference models operate based on human selected features and traditional machine learning algorithms. However, human selected features are sensitive to changes in traffic and environmental conditions and susceptible to personal biases, which can make them inefficient. One way to overcome these problems is to use neural networks capable of extracting high-level features from raw input. In this study, the convolutional neural network (CNN) architecture is used to predict the trip mode distribution based on raw GPS trajectory data. The key innovation of this paper is the design of the layout of the input layer of CNN as well as normalization operation, in a way that is not only compatible with the CNN architecture but can also represent the fundamental features of motion including speed, acceleration, jerk, and Bearing rate. The highest prediction accuracy achieved with the proposed configuration for the convolutional neural network with batch normalization is 85.26%.

Keywords: predicting, deep learning, neural network, urban trip

Procedia PDF Downloads 137

Authors: Tayyibah Khanam, Mohammad Saad Alam, Sanchari Deb, Yasser Rafat

Abstract:

Electric vehicles are considered as viable replacements to gasoline cars since they help in reducing harmful emissions and stimulate power generation through renewable energy sources, hence contributing to sustainability. However, one of the significant obstacles in the mass deployment of electric vehicles is the charging time anxiety among users and, thus, the subsequent large waiting times for available chargers at charging stations. Data analytics, on the other hand, has revolutionized the decision-making tasks of management and operating systems since its arrival. In this paper, we attempt to optimize the choice of EV charging stations for users in their vicinity by minimizing the time taken to reach the charging stations and the waiting times for available chargers. Time taken to travel to the charging station is calculated by the Google Maps API and the waiting times are predicted by polynomial regression of the historical data stored. The proposed framework utilizes real-time data and historical data from all operating charging stations in the city and assists the user in finding the best suitable charging station for their current situation and can be implemented in a mobile phone application. The algorithm successfully predicts the most optimal choice of a charging station and the minimum required time for various sample data sets.

Keywords: charging data, electric vehicles, machine learning, waiting times

Procedia PDF Downloads 191

Authors: Vidhu Mitha Goutham, Ruma Mukherjee

Abstract:

The low availability of well-trained, unlimited, dynamic-access models for specific languages makes it hard for corporate users to adopt quick translation techniques and incorporate them into product solutions. As translation tasks increasingly require a dynamic sequence learning curve; stable, cost-free opensource models are scarce. We survey and compare current translation techniques and propose a modified sequence to sequence model repurposed with attention techniques. Sequence learning using an encoder-decoder model is now paving the path for higher precision levels in translation. Using a Convolutional Neural Network (CNN) encoder and a Recurrent Neural Network (RNN) decoder background, we use Fairseq tools to produce an end-to-end bilingually trained Spanish-English machine translation model including source language detection. We acquire competitive results using a duo-lingo-corpus trained model to provide for prospective, ready-made plug-in use for compound sentences and document translations. Our model serves a decent system for large, organizational data translation needs. While acknowledging its shortcomings and future scope, it also identifies itself as a well-optimized deep neural network model and solution.

Keywords: attention, encoder-decoder, Fairseq, Seq2Seq, Spanish, translation

Procedia PDF Downloads 174

Authors: Ntaganda J.M., Maniraguha J.D., Mukeshimana S., Harelimana D, Bizimungu T., Ruataganda E.

Abstract:

This paper presents the design of Graphic User Interface (GUI) in Matlab as interaction tool between human and machine. The designed GUI can be used by medical doctors and other experts particularly the physiologists. Matlab packages and estimated parameters of the mathematical model of cardiovascular-respiratory system developed in Rwandan context are used in GUI. The ordinary differential equations (ODE’s) govern a mathematical model in designing GUI in Matlab and a window that sets model estimated parameters and the measured parameters by any user. For healthy subject, these measured parameters include heart rate, systolic blood and diastolic blood pressure, partial pressure of oxygen in arterial blood, partial pressure of carbon dioxide in arterial blood, concentration of bound and dissolved oxygen in the mixed venous blood entering the lungs, and concentration of bound and dissolved carbon dioxide in the mixed venous blood entering the lungs. The results of numerical test give a consistent appearance as empirically known results.

Keywords: Graphic User Interface, mathematical model, cardiovascur-respiratory system, walking physical activity, blood pressure, oxygen

Procedia PDF Downloads 117

Authors: Adrian-Gabriel Chifu, Sebastien Fournier

Abstract:

One of the challenges of natural language understanding is to deal with the subjectivity of sentences, which may express opinions and emotions that add layers of complexity and nuance. Sentiment analysis is a field that aims to extract and analyze these subjective elements from text, and it can be applied at different levels of granularity, such as document, paragraph, sentence, or aspect. Aspect-based sentiment analysis is a well-studied topic with many available data sets and models. However, there is no clear definition of what makes a sentence difficult for aspect-based sentiment analysis. In this paper, we explore this question by conducting an experiment with three data sets: ”Laptops”, ”Restaurants”, and ”MTSC” (Multi-Target-dependent Sentiment Classification), and a merged version of these three datasets. We study the impact of domain diversity and syntactic diversity on difficulty. We use a combination of classifiers to identify the most difficult sentences and analyze their characteristics. We employ two ways of defining sentence difficulty. The first one is binary and labels a sentence as difficult if the classifiers fail to correctly predict the sentiment polarity. The second one is a six-level scale based on how many of the top five best-performing classifiers can correctly predict the sentiment polarity. We also define 9 linguistic features that, combined, aim at estimating the difficulty at sentence level.

Keywords: sentiment analysis, difficulty, classification, machine learning

Procedia PDF Downloads 86

Authors: Masaumi Nakahara, Sou Watanabe, Hiromichi Ogi, Atsuhiro Shibata, Kazunori Nomura

Abstract:

In the Japan Atomic Energy Agency, three types of experimental research, advanced reactor fuel reprocessing, radioactive waste disposal, and nuclear fuel cycle technology, have been carried out at the Chemical Processing Facility. The facility has generated high level radioactive liquid and solid wastes in hot cells. The high level radioactive solid waste is divided into three main categories, a flammable waste, a non-flammable waste, and a solid reagent waste. A plastic product is categorized into the flammable waste and molten with a heating mantle. The non-flammable waste is cut with a band saw machine for reducing the volume. Among the solid reagent waste, a used adsorbent after the experiments is heated, and an extractant is decomposed for its stabilization. All high level radioactive solid wastes in the hot cells are packed in a high level radioactive solid waste can. The high level radioactive solid waste can is transported to the 2nd High Active Solid Waste Storage in the Tokai Reprocessing Plant in the Japan Atomic Energy Agency.

Keywords: high level radioactive solid waste, advanced reactor fuel reprocessing, radioactive waste disposal, nuclear fuel cycle technology

Procedia PDF Downloads 155

Authors: Wafaa Salah, Mostafa A. Salama, Jane Doe

Abstract:

Recently, the corporate social performance (CSP) is not less important than the corporate financial performance (CFP). Debate still exists about the nature of the relationship between the CSP and CFP, whether it is a positive, negative or a neutral correlation. The objective of this study is to explore the relationship between corporate social responsibility (CSR) reports and CFP. The study uses the accounting-based and market-based quantitative measures to quantify the financial performance of seven organizations listed on the Egyptian Stock Exchange in 2007-2014. Then uses the information retrieval technologies to quantify the contribution of each of the three dimensions of the corporate social responsibility report (environmental, social and economic). Finally, the correlation between these two sets of variables is viewed together in a model to detect the correlations between them. This model is applied on seven firms that generate social responsibility reports. The results show a positive correlation between the Earnings per share (market based measure) and the economical dimension in the CSR report. On the other hand, total assets and property, plant and equipment (accounting-based measure) are positively correlated to the environmental and social dimensions of the CSR reports. While there is not any significant relationship between ROA, ROE, Operating income and corporate social responsibility. This study contributes to the literature by providing more clarification of the relationship between CFP and the isolated CSR activities in a developing country.

Keywords: financial, social, machine learning, corporate social performance, corporate social responsibility

Procedia PDF Downloads 310

Authors: Rafi Ullah, Shah Muhammad Emaduddin, Taha Jilani

Abstract:

Vehicle Adhoc Networks (VANET) belongs to a special class of Mobile Adhoc Network (MANET) with high mobility. Network is created by road side vehicles equipped with communication devices like GPS and Wifi etc. Since the environment is highly dynamic due to difference in speed and high mobility of vehicles and weak stability of the network connection, it is a challenging task to design an efficient routing protocol for such an unstable environment. Our proposed algorithm uses heuristic for the calculation of optimal path for routing the packet efficiently in collaboration with several other parameters like geographical location, speed, priority, the distance among the vehicles, communication range, and networks congestion. We have incorporated probabilistic, heuristic and machine learning based approach inconsistency with the relay function of the memory buffer to keep the packet moving towards the destination. These parameters when used in collaboration provide us a very strong and admissible heuristics. We have mathematically proved that the proposed technique is efficient for the routing of packets, especially in a medical emergency situation. These networks can be used for medical emergency, security, entertainment and routing purposes.

Keywords: heuristics routing, intelligent routing, VANET, route optimization

Procedia PDF Downloads 176

Authors: Vimala Rani P, Narasimha Malikarjunan, Mercy Shalinie S

Abstract:

Data Networking was brought forth as an instantiation of information-centric networking. The attackers can send a colossal number of spoofs to take hold of the Pending Interest Table (PIT) named an Interest Flooding attack (IFA) since the in- interests are recorded in the PITs of the intermediate routers until they receive corresponding Data Packets are go beyond the time limit. These attacks can be detrimental to network performance. PIT expiration rate or the Interest satisfaction rate, which cannot differentiate the IFA from attacks, is the criterion Traditional IFA detection techniques are concerned with. Threshold values can casually affect Threshold-based traditional methods. This article proposes an accurate IFA detection mechanism based on a Multiple Feature-based Extreme Learning Machine (MF-ELM). Accuracy of the attack detection can be increased by presenting the entropy of Internet names, Interest satisfaction rate and PIT usage as features extracted in the MF-ELM classifier. Furthermore, we deploy a queue-based hostile Interest prefix mitigation mechanism. The inference of this real-time test bed is that the mechanism can help the network to resist IFA with higher accuracy and efficiency.

Keywords: information-centric network, pending interest table, interest flooding attack, MF-ELM classifier, queue-based mitigation strategy

Procedia PDF Downloads 204

Authors: Taylor Kolody, Farkhund Iqbal, Rabia Batool, Benjamin Fung, Mohammed Hussaeni, Saiqa Aleem

Abstract:

The rising threat of climate change has led to an increase in public awareness and care about our collective and individual environmental impact. A key component of this impact is our use of cars and other polluting forms of transportation, but it is often difficult for an individual to know how severe this impact is. While there are applications that offer this feedback, they require manual entry of what transportation mode was used for a given trip, which can be burdensome. In order to alleviate this shortcoming, a data from the 2016 TRIPlab datasets has been used to train a variety of machine learning models to automatically recognize the mode of transportation. The accuracy of 89.6% is achieved using single deep neural network model with Gated Recurrent Unit (GRU) architecture applied directly to trip data points over 4 primary classes, namely walking, public transit, car, and bike. These results are comparable in accuracy to results achieved by others using ensemble methods and require far less computation when classifying new trips. The lack of trip context data, e.g., bus routes, bike paths, etc., and the need for only a single set of weights make this an appropriate methodology for applications hoping to reach a broad demographic and have responsive feedback.

Keywords: classification, gated recurrent unit, recurrent neural network, transportation

Procedia PDF Downloads 136

Authors: Zhou Xiong, Kang Shao Bo, Yang Bo

Abstract:

To date, high-performance structural steel has been widely used for columns in construction practices due to its significant advantages over conventional steel. However, the same design approach with conventional steel columns is still adopted in the design of high-performance steel columns. As a result, its superior properties cannot be fully considered in design. This paper conducts a test and finite element analysis on the overall stability behaviour of welded Q460GJ steel box columns. In the test, four steel columns with different slenderness and width-to-thickness ratio were compressed under an axial compression testing machine. And finite element models were established in which material nonlinearity and residual stress distributions of test columns were included. Then, comparisons were made between test results and finite element result, it showed that finite element analysis results are agree well with the test result. It means that the test and finite element model are reliable. Then, we compared the test result with the design value calculated by current code, the result showed that Q460GJ steel box columns have the higher overall buckling capacity than the design value. It is necessary to update the design curves for Q460GJ steel columns so that the overall stability capacity of Q460GJ box columns can be designed appropriately.

Keywords: axial compression, box columns, global buckling, numerical simulations, Q460GJ steel

Procedia PDF Downloads 403

Authors: B. Soni, S. Biswas

Abstract:

Metal foams of Al find diverse applications in several industrial sectors such as in automotive and sports equipment industry as impact, acoustic and vibration absorbers, the aerospace industry as structural components in turbines and spatial cones, in the naval industry as low frequency vibration absorbers, and in construction industry as sound barriers inside tunnels, as fire proof materials and structure protection systems against explosions and even in heat exchangers, orthopedic components, and decorative items. Here, we report on the development of Al foams by a low cost and convenient technique of salt replication method with efficient control over size, geometry and distribution of the pores. Sodium bicarbonate was used as the foaming agent to form the porous refractory salt pattern. The mixed refractory salt slurry was microwave dried followed by sintering for selected time periods. Molten Al was infiltrated into the salt pattern in an inert atmosphere at a pressure of 2 bars. The final products were obtained by leaching out the refractory salt pattern. Mechanical properties of the derived samples were studied with a universal testing machine. The results were analyzed in correlation with their microstructural features evaluated with a scanning electron microscope (SEM).

Keywords: metal foam, Al, salt replication method, mechanical properties, SEM

Procedia PDF Downloads 350

Authors: Shinoy Vengaramkode Bhaskaran, Kaushik Sathupadi, Sandesh Achar

Abstract:

Human locomotion recognition is important in a variety of sectors, such as robotics, security, healthcare, fitness tracking and cloud computing. With the increasing pervasiveness of peripheral devices, particularly Inertial Measurement Units (IMUs) sensors, researchers have attempted to exploit these advancements in order to precisely and efficiently identify and categorize human activities. This research paper introduces a state-of-the-art methodology for the recognition of human locomotion patterns in a cloud environment. The methodology is based on a publicly available benchmark dataset. The investigation implements a denoising and windowing strategy to deal with the unprocessed data. Next, feature extraction is adopted to abstract the main cues from the data. The SelectKBest strategy is used to abstract optimal features from the data. Furthermore, state-of-the-art ML classifiers are used to evaluate the performance of the system, including logistic regression, random forest, gradient boosting and SVM have been investigated to accomplish precise locomotion classification. Finally, a detailed comparative analysis of results is presented to reveal the performance of recognition models.

Keywords: artificial intelligence, cloud computing, IoT, human locomotion, gradient boosting, random forest, neural networks, body-worn sensors

Procedia PDF Downloads 7

Authors: Surajit Chakrabarty, Piyush Chauhan, Subhasis Panda, Sujoy Bhattacharya

Abstract:

A large proportion of roads in India are not well maintained as per the laid down public safety guidelines leading to loss of direction control and fatal accidents. We propose a technique to detect speed breakers and potholes using mobile sensor data captured from multiple vehicles and provide a profile of the road. This would, in turn, help in monitoring roads and revolutionize digital maps. Incorporating randomness in the model formulation for detection of speed breakers and potholes is crucial due to substantial heterogeneity observed in data obtained using a mobile application from multiple vehicles driven by different drivers. This is accomplished with Hidden Markov Models, whose hidden state sequence is found for each time step given the observables sequence, and are then fed as input to LSTM network with peephole connections. A precision score of 0.96 and 0.63 is obtained for classifying bumps and potholes, respectively, a significant improvement from the machine learning based models. Further visualization of bumps/potholes is done by converting time series to images using Markov Transition Fields where a significant demarcation among bump/potholes is observed.

Keywords: deep learning, hidden Markov model, pothole, speed breaker

Procedia PDF Downloads 143

Authors: Lamis Naddaf, Yuval Tabach

Abstract:

In this research, we identified the missense genetic variants that have the potential to enhance resistance against cancer. Such field has not been widely explored, as researchers tend to investigate mutations that cause diseases, in response to the suffering of patients, rather than those mutations that protect from them. In conjunction with the genomic revolution, and the advances in genetic engineering and synthetic biology, identifying the protective variants will increase the power of genotype-phenotype predictions and can have significant implications on improved risk estimation, diagnostics, prognosis and even for personalized therapy and drug discovery. To approach our goal, we systematically investigated the sites of the coding genomes and picked up the alleles that showed a correlation with the species’ cancer resistance. We predicted 250 protecting variants (PVs) with a 0.01 false discovery rate and more than 20 thousand PVs with a 0.25 false discovery rate. Cancer resistance in Mammals and reptiles was significantly predicted by the number of PVs a species has. Moreover, Genes enriched with the protecting variants are enriched in pathways relevant to tumor suppression like pathways of Hedgehog signaling and silencing, which its improper activation is associated with the most common form of cancer malignancy. We also showed that the PVs are more abundant in healthy people compared to cancer patients within different human races.

Keywords: comparative genomics, machine learning, cancer resistance, cancer-protecting alleles

Procedia PDF Downloads 95

Authors: Shota Nagata, Kazuya Okubo, Toru Fujii

Abstract:

The purpose of this study is to investigate the proper condition in extracting carbon fibers as the reinforcement of composite molded by injection method. Recycled carbon fibers were extracted from wasted CFRP by pyrolyzing epoxy matrix of CFRP under air atmosphere at different temperature conditions 400, 600 and 800°C in this study. Recycled carbon fiber reinforced polypropylene (RCF/PP) pellets were prepared using twin screw extruder. The RCF/PP specimens were molded into dumbbell shaped specimens using injection molding machine. The tensile strength of recycled carbon fiber was decreased with rising pyrolysis temperature from 400 to 800°C. However, superior mechanical properties of tensile strength, tensile modulus and fracture strain of RCF/PP specimen were obtained when the extracting temperature was 600°C. Almost fibers in RCF/PP specimens were aligned in the mold filling direction in this study when the extracting temperature was 600°C. To discuss the results, the failure mechanisms of RCF/PP specimens was shown schematically. Finally, it was concluded that the temperature condition at 600°C should be selected in extracting carbon fibers as the reinforcement of RCF/PP composite molded by injection method.

Keywords: CFRP, recycled carbon fiber, injection molding, mechanical properties, fiber orientation, failure mechanism

Procedia PDF Downloads 444

Authors: Justin Tang

Abstract:

Balance tests are frequently used to diagnose concussions on the sidelines of sporting events. Manual scoring, however, is labor intensive and subjective, and many concussions go undetected. This study institutes a novel approach to conducting the Balance Error Scoring System (BESS) more quantitatively using Microsoft’s gaming system Kinect, which uses a contactless sensor and several cameras to receive data and estimate body limb positions. Using a machine learning approach, Visual Gesture Builder, and a deterministic approach, MATLAB, we tested whether the Kinect can differentiate between “correct” and erroneous stances of the BESS. We created the two separate solutions by recording test videos to teach the Kinect correct stances and by developing a code using Java. Twenty-two subjects were asked to perform a series of BESS tests while the Kinect was collecting data. The Kinect recorded the subjects and mapped key joints onto their bodies to obtain angles and measurements that are interpreted by the software. Through VGB and MATLAB, the videos are analyzed to enumerate the number of errors committed during testing. The resulting statistics demonstrate a high correlation between manual scoring and the Kinect approaches, indicating the viability of the use of remote tracking devices in conducting concussion tests.

Keywords: automated, concussion detection, contactless sensors, microsoft kinect

Procedia PDF Downloads 316