Abstracts | Computer and Systems Engineering
Commenced in January 2007
Frequency: Monthly
Edition: International
Paper Count: 415

World Academy of Science, Engineering and Technology

[Computer and Systems Engineering]

Online ISSN : 1307-6892

295 Classification of Generative Adversarial Network Generated Multivariate Time Series Data Featuring Transformer-Based Deep Learning Architecture

Authors: Thrivikraman Aswathi, S. Advaith

Abstract:

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

Keywords: GAN, transformer, classification, multivariate time series

Procedia PDF Downloads 123
294 Performance Evaluation of the Classic seq2seq Model versus a Proposed Semi-supervised Long Short-Term Memory Autoencoder for Time Series Data Forecasting

Authors: Aswathi Thrivikraman, S. Advaith

Abstract:

The study is aimed at designing encoders for deciphering intricacies in time series data by redescribing the dynamics operating on a lower-dimensional manifold. A semi-supervised LSTM autoencoder is devised and investigated to see if the latent representation of the time series data can better forecast the data. End-to-end training of the LSTM autoencoder, together with another LSTM network that is connected to the latent space, forces the hidden states of the encoder to represent the most meaningful latent variables relevant for forecasting. Furthermore, the study compares the predictions with those of a traditional seq2seq model.

Keywords: LSTM, autoencoder, forecasting, seq2seq model

Procedia PDF Downloads 148
293 Deep Learning for Qualitative and Quantitative Grain Quality Analysis Using Hyperspectral Imaging

Authors: Ole-Christian Galbo Engstrøm, Erik Schou Dreier, Birthe Møller Jespersen, Kim Steenstrup Pedersen

Abstract:

Grain quality analysis is a multi-parameterized problem that includes a variety of qualitative and quantitative parameters such as grain type classification, damage type classification, and nutrient regression. Currently, these parameters require human inspection, a multitude of instruments employing a variety of sensor technologies, and predictive model types or destructive and slow chemical analysis. This paper investigates the feasibility of applying near-infrared hyperspectral imaging (NIR-HSI) to grain quality analysis. For this study two datasets of NIR hyperspectral images in the wavelength range of 900 nm - 1700 nm have been used. Both datasets contain images of sparsely and densely packed grain kernels. The first dataset contains ~87,000 image crops of bulk wheat samples from 63 harvests where protein value has been determined by the FOSS Infratec NOVA which is the golden industry standard for protein content estimation in bulk samples of cereal grain. The second dataset consists of ~28,000 image crops of bulk grain kernels from seven different wheat varieties and a single rye variety. In the first dataset, protein regression analysis is the problem to solve while variety classification analysis is the problem to solve in the second dataset. Deep convolutional neural networks (CNNs) have the potential to utilize spatio-spectral correlations within a hyperspectral image to simultaneously estimate the qualitative and quantitative parameters. CNNs can autonomously derive meaningful representations of the input data reducing the need for advanced preprocessing techniques required for classical chemometric model types such as artificial neural networks (ANNs) and partial least-squares regression (PLS-R). A comparison between different CNN architectures utilizing 2D and 3D convolution is conducted. These results are compared to the performance of ANNs and PLS-R. Additionally, a variety of preprocessing techniques from image analysis and chemometrics are tested. These include centering, scaling, standard normal variate (SNV), Savitzky-Golay (SG) filtering, and detrending. The results indicate that the combination of NIR-HSI and CNNs has the potential to be the foundation for an automatic system unifying qualitative and quantitative grain quality analysis within a single sensor technology and predictive model type.

Keywords: deep learning, grain analysis, hyperspectral imaging, preprocessing techniques

Procedia PDF Downloads 93
292 Recent Developments in the Application of Deep Learning to Stock Market Prediction

Authors: Shraddha Jain Sharma, Ratnalata Gupta

Abstract:

Predicting stock movements in the financial market is both difficult and rewarding. Analysts and academics are increasingly using advanced approaches such as machine learning techniques to anticipate stock price patterns, thanks to the expanding capacity of computing and the recent advent of graphics processing units and tensor processing units. Stock market prediction is a type of time series prediction that is incredibly difficult to do since stock prices are influenced by a variety of financial, socioeconomic, and political factors. Furthermore, even minor mistakes in stock market price forecasts can result in significant losses for companies that employ the findings of stock market price prediction for financial analysis and investment. Soft computing techniques are increasingly being employed for stock market prediction due to their better accuracy than traditional statistical methodologies. The proposed research looks at the need for soft computing techniques in stock market prediction, the numerous soft computing approaches that are important to the field, past work in the area with their prominent features, and the significant problems or issue domain that the area involves. For constructing a predictive model, the major focus is on neural networks and fuzzy logic. The stock market is extremely unpredictable, and it is unquestionably tough to correctly predict based on certain characteristics. This study provides a complete overview of the numerous strategies investigated for high accuracy prediction, with a focus on the most important characteristics.

Keywords: stock market prediction, artificial intelligence, artificial neural networks, fuzzy logic, accuracy, deep learning, machine learning, stock price, trading volume

Procedia PDF Downloads 85
291 Pivoting to Fortify our Digital Self: Revealing the Need for Personal Cyber Insurance

Authors: Richard McGregor, Carmen Reaiche, Stephen Boyle

Abstract:

Cyber threats are a relatively recent phenomenon and offer cyber insurers a dynamic and intelligent peril. As individuals en mass become increasingly digitally dependent, Personal Cyber Insurance (PCI) offers an attractive option to mitigate cyber risk at a personal level. This abstract proposes a literature review that conceptualises a framework for siting Personal Cyber Insurance (PCI) within the context of cyberspace. The lack of empirical research within this domain demonstrates an immediate need to define the scope of PCI to allow cyber insurers to understand personal cyber risk threats and vectors, customer awareness, capabilities, and their associated needs. Additionally, this will allow cyber insurers to conceptualise appropriate frameworks allowing effective management and distribution of PCI products and services within a landscape often in-congruent with risk attributes commonly associated with traditional personal line insurance products. Cyberspace has provided significant improvement to the quality of social connectivity and productivity during past decades and allowed enormous capability uplift of information sharing and communication between people and communities. Conversely, personal digital dependency furnish ample opportunities for adverse cyber events such as data breaches and cyber-attacksthus introducing a continuous and insidious threat of omnipresent cyber risk–particularly since the advent of the COVID-19 pandemic and wide-spread adoption of ‘work-from-home’ practices. Recognition of escalating inter-dependencies, vulnerabilities and inadequate personal cyber behaviours have prompted efforts by businesses and individuals alike to investigate strategies and tactics to mitigate cyber risk – of which cyber insurance is a viable, cost-effective option. It is argued that, ceteris parabus, the nature of cyberspace intrinsically provides characteristic peculiarities that pose significant and bespoke challenges to cyber insurers, often in-congruent with risk attributes commonly associated with traditional personal line insurance products. These challenges include (inter alia) a paucity of historical claim/loss data for underwriting and pricing purposes, interdependencies of cyber architecture promoting high correlation of cyber risk, difficulties in evaluating cyber risk, intangibility of risk assets (such as data, reputation), lack of standardisation across the industry, high and undetermined tail risks, and moral hazard among others. This study proposes a thematic overview of the literature deemed necessary to conceptualise the challenges to issuing personal cyber coverage. There is an evident absence of empirical research appertaining to PCI and the design of operational business models for this business domain, especially qualitative initiatives that (1) attempt to define the scope of the peril, (2) secure an understanding of the needs of both cyber insurer and customer, and (3) to identify elements pivotal to effective management and profitable distribution of PCI - leading to an argument proposed by the author that postulates that the traditional general insurance customer journey and business model are ill-suited for the lineaments of cyberspace. The findings of the review confirm significant gaps in contemporary research within the domain of personal cyber insurance.

Keywords: cyberspace, personal cyber risk, personal cyber insurance, customer journey, business model

Procedia PDF Downloads 96
290 Improving Activity Recognition Classification of Repetitious Beginner Swimming Using a 2-Step Peak/Valley Segmentation Method with Smoothing and Resampling for Machine Learning

Authors: Larry Powell, Seth Polsley, Drew Casey, Tracy Hammond

Abstract:

Human activity recognition (HAR) systems have shown positive performance when recognizing repetitive activities like walking, running, and sleeping. Water-based activities are a reasonably new area for activity recognition. However, water-based activity recognition has largely focused on supporting the elite and competitive swimming population, which already has amazing coordination and proper form. Beginner swimmers are not perfect, and activity recognition needs to support the individual motions to help beginners. Activity recognition algorithms are traditionally built around short segments of timed sensor data. Using a time window input can cause performance issues in the machine learning model. The window’s size can be too small or large, requiring careful tuning and precise data segmentation. In this work, we present a method that uses a time window as the initial segmentation, then separates the data based on the change in the sensor value. Our system uses a multi-phase segmentation method that pulls all peaks and valleys for each axis of an accelerometer placed on the swimmer’s lower back. This results in high recognition performance using leave-one-subject-out validation on our study with 20 beginner swimmers, with our model optimized from our final dataset resulting in an F-Score of 0.95.

Keywords: time window, peak/valley segmentation, feature extraction, beginner swimming, activity recognition

Procedia PDF Downloads 111
289 JaCoText: A Pretrained Model for Java Code-Text Generation

Authors: Jessica Lopez Espejel, Mahaman Sanoussi Yahaya Alassan, Walid Dahhane, El Hassane Ettifouri

Abstract:

Pretrained transformer-based models have shown high performance in natural language generation tasks. However, a new wave of interest has surged: automatic programming language code generation. This task consists of translating natural language instructions to a source code. Despite the fact that well-known pre-trained models on language generation have achieved good performance in learning programming languages, effort is still needed in automatic code generation. In this paper, we introduce JaCoText, a model based on Transformer neural network. It aims to generate java source code from natural language text. JaCoText leverages the advantages of both natural language and code generation models. More specifically, we study some findings from state of the art and use them to (1) initialize our model from powerful pre-trained models, (2) explore additional pretraining on our java dataset, (3) lead experiments combining the unimodal and bimodal data in training, and (4) scale the input and output length during the fine-tuning of the model. Conducted experiments on CONCODE dataset show that JaCoText achieves new state-of-the-art results.

Keywords: java code generation, natural language processing, sequence-to-sequence models, transformer neural networks

Procedia PDF Downloads 273
288 A User-Directed Approach to Optimization via Metaprogramming

Authors: Eashan Hatti

Abstract:

In software development, programmers often must make a choice between high-level programming and high-performance programs. High-level programming encourages the use of complex, pervasive abstractions. However, the use of these abstractions degrades performance-high performance demands that programs be low-level. In a compiler, the optimizer attempts to let the user have both. The optimizer takes high-level, abstract code as an input and produces low-level, performant code as an output. However, there is a problem with having the optimizer be a built-in part of the compiler. Domain-specific abstractions implemented as libraries are common in high-level languages. As a language’s library ecosystem grows, so does the number of abstractions that programmers will use. If these abstractions are to be performant, the optimizer must be extended with new optimizations to target them, or these abstractions must rely on existing general-purpose optimizations. The latter is often not as effective as needed. The former presents too significant of an effort for the compiler developers, as they are the only ones who can extend the language with new optimizations. Thus, the language becomes more high-level, yet the optimizer – and, in turn, program performance – falls behind. Programmers are again confronted with a choice between high-level programming and high-performance programs. To investigate a potential solution to this problem, we developed Peridot, a prototype programming language. Peridot’s main contribution is that it enables library developers to easily extend the language with new optimizations themselves. This allows the optimization workload to be taken off the compiler developers’ hands and given to a much larger set of people who can specialize in each problem domain. Because of this, optimizations can be much more effective while also being much more numerous. To enable this, Peridot supports metaprogramming designed for implementing program transformations. The language is split into two fragments or “levels”, one for metaprogramming, the other for high-level general-purpose programming. The metaprogramming level supports logic programming. Peridot’s key idea is that optimizations are simply implemented as metaprograms. The meta level supports several specific features which make it particularly suited to implementing optimizers. For instance, metaprograms can automatically deduce equalities between the programs they are optimizing via unification, deal with variable binding declaratively via higher-order abstract syntax, and avoid the phase-ordering problem via non-determinism. We have found that this design centered around logic programming makes optimizers concise and easy to write compared to their equivalents in functional or imperative languages. Overall, implementing Peridot has shown that its design is a viable solution to the problem of writing code which is both high-level and performant.

Keywords: optimization, metaprogramming, logic programming, abstraction

Procedia PDF Downloads 81
287 Conflict Resolution in Fuzzy Rule Base Systems Using Temporal Modalities Inference

Authors: Nasser S. Shebka

Abstract:

Fuzzy logic is used in complex adaptive systems where classical tools of representing knowledge are unproductive. Nevertheless, the incorporation of fuzzy logic, as it’s the case with all artificial intelligence tools, raised some inconsistencies and limitations in dealing with increased complexity systems and rules that apply to real-life situations and hinders the ability of the inference process of such systems, but it also faces some inconsistencies between inferences generated fuzzy rules of complex or imprecise knowledge-based systems. The use of fuzzy logic enhanced the capability of knowledge representation in such applications that requires fuzzy representation of truth values or similar multi-value constant parameters derived from multi-valued logic, which set the basis for the three t-norms and their based connectives which are actually continuous functions and any other continuous t-norm can be described as an ordinal sum of these three basic ones. However, some of the attempts to solve this dilemma were an alteration to fuzzy logic by means of non-monotonic logic, which is used to deal with the defeasible inference of expert systems reasoning, for example, to allow for inference retraction upon additional data. However, even the introduction of non-monotonic fuzzy reasoning faces a major issue of conflict resolution for which many principles were introduced, such as; the specificity principle and the weakest link principle. The aim of our work is to improve the logical representation and functional modelling of AI systems by presenting a method of resolving existing and potential rule conflicts by representing temporal modalities within defeasible inference rule-based systems. Our paper investigates the possibility of resolving fuzzy rules conflict in a non-monotonic fuzzy reasoning-based system by introducing temporal modalities and Kripke's general weak modal logic operators in order to expand its knowledge representation capabilities by means of flexibility in classifying newly generated rules, and hence, resolving potential conflicts between these fuzzy rules. We were able to address the aforementioned problem of our investigation by restructuring the inference process of the fuzzy rule-based system. This is achieved by using time-branching temporal logic in combination with restricted first-order logic quantifiers, as well as propositional logic to represent classical temporal modality operators. The resulting findings not only enhance the flexibility of complex rule-base systems inference process but contributes to the fundamental methods of building rule bases in such a manner that will allow for a wider range of applicable real-life situations derived from a quantitative and qualitative knowledge representational perspective.

Keywords: fuzzy rule-based systems, fuzzy tense inference, intelligent systems, temporal modalities

Procedia PDF Downloads 85
286 Closest Possible Neighbor of a Different Class: Explaining a Model Using a Neighbor Migrating Generator

Authors: Hassan Eshkiki, Benjamin Mora

Abstract:

The Neighbor Migrating Generator is a simple and efficient approach to finding the closest potential neighbor(s) with a different label for a given instance and so without the need to calibrate any kernel settings at all. This allows determining and explaining the most important features that will influence an AI model. It can be used to either migrate a specific sample to the class decision boundary of the original model within a close neighborhood of that sample or identify global features that can help localising neighbor classes. The proposed technique works by minimizing a loss function that is divided into two components which are independently weighted according to three parameters α, β, and ω, α being self-adjusting. Results show that this approach is superior to past techniques when detecting the smallest changes in the feature space and may also point out issues in models like over-fitting.

Keywords: explainable AI, EX AI, feature importance, counterfactual explanations

Procedia PDF Downloads 170
285 Minimizing Mutant Sets by Equivalence and Subsumption

Authors: Samia Alblwi, Amani Ayad

Abstract:

Mutation testing is the art of generating syntactic variations of a base program and checking whether a candidate test suite can identify all the mutants that are not semantically equivalent to the base: this technique is widely used by researchers to select quality test suites. One of the main obstacles to the widespread use of mutation testing is cost: even small pro-grams (a few dozen lines of code) can give rise to a large number of mutants (up to hundreds): this has created an incentive to seek to reduce the number of mutants while preserving their collective effectiveness. Two criteria have been used to reduce the size of mutant sets: equiva-lence, which aims to partition the set of mutants into equivalence classes modulo semantic equivalence, and selecting one representative per class; subsumption, which aims to define a partial ordering among mutants that ranks mutants by effectiveness and seeks to select maximal elements in this ordering. In this paper we analyze these two policies using analytical and em-pirical criteria.

Keywords: mutation testing, mutant sets, mutant equivalence, mutant subsumption, mutant set minimization

Procedia PDF Downloads 56
284 Learning the History of a Tuscan Village: A Serious Game Using Geolocation Augmented Reality

Authors: Irene Capecchi, Tommaso Borghini, Iacopo Bernetti

Abstract:

An important tool for the enhancement of cultural sites is serious games (SG), i.e., games designed for educational purposes; SG is applied in cultural sites through trivia, puzzles, and mini-games for participation in interactive exhibitions, mobile applications, and simulations of past events. The combination of Augmented Reality (AR) and digital cultural content has also produced examples of cultural heritage recovery and revitalization around the world. Through AR, the user perceives the information of the visited place in a more real and interactive way. Another interesting technological development for the revitalization of cultural sites is the combination of AR and Global Positioning System (GPS), which integrated have the ability to enhance the user's perception of reality by providing historical and architectural information linked to specific locations organized on a route. To the author’s best knowledge, there are currently no applications that combine GPS AR and SG for cultural heritage revitalization. The present research focused on the development of an SG based on GPS and AR. The study area is the village of Caldana in Tuscany, Italy. Caldana is a fortified Renaissance village; the most important architectures are the walls, the church of San Biagio, the rectory, and the marquis' palace. The historical information is derived from extensive research by the Department of Architecture at the University of Florence. The storyboard of the SG is based on the history of the three characters who built the village: marquis Marcello Agostini, who was commissioned by Cosimo I de Medici, Grand Duke of Tuscany, to build the village, his son Ippolito and his architect Lorenzo Pomarelli. The three historical characters were modeled in 3D using the freeware MakeHuman and imported into Blender and Mixamo to associate a skeleton and blend shapes to have gestural animations and reproduce lip movement during speech. The Unity Rhubarb Lip Syncer plugin was used for the lip sync animation. The historical costumes were created by Marvelous Designer. The application was developed using the Unity 3D graphics and game engine. The AR+GPS Location plugin was used to position the 3D historical characters based on GPS coordinates. The ARFoundation library was used to display AR content. The SG is available in two versions: for children and adults. the children's version consists of finding a digital treasure consisting of valuable items and historical rarities. Players must find 9 village locations where 3D AR models of historical figures explaining the history of the village provide clues. To stimulate players, there are 3 levels of rewards for every 3 clues discovered. The rewards consist of AR masks for archaeologist, professor, and explorer. At the adult level, the SG consists of finding the 16 historical landmarks in the village, and learning historical and architectural information interactively and engagingly. The application is being tested on a sample of adults and children. Test subjects will be surveyed on a Likert scale to find out their perceptions of using the app and the learning experience between the guided tour and interaction with the app.

Keywords: augmented reality, cultural heritage, GPS, serious game

Procedia PDF Downloads 88
283 Training for Safe Tree Felling in the Forest with Symmetrical Collaborative Virtual Reality

Authors: Irene Capecchi, Tommaso Borghini, Iacopo Bernetti

Abstract:

One of the most common pieces of equipment still used today for pruning, felling, and processing trees is the chainsaw in forestry. However, chainsaw use highlights dangers and one of the highest rates of accidents in both professional and non-professional work. Felling is proportionally the most dangerous phase, both in severity and frequency, because of the risk of being hit by the plant the operator wants to cut down. To avoid this, a correct sequence of chainsaw cuts must be taught concerning the different conditions of the tree. Virtual reality (VR) makes it possible to virtually simulate chainsaw use without danger of injury. The limitations of the existing applications are as follow. The existing platforms are not symmetrical collaborative because the trainee is only in virtual reality, and the trainer can only see the virtual environment on a laptop or PC, and this results in an inefficient teacher-learner relationship. Therefore, most applications only involve the use of a virtual chainsaw, and the trainee thus cannot feel the real weight and inertia of a real chainsaw. Finally, existing applications simulate only a few cases of tree felling. The objectives of this research were to implement and test a symmetrical collaborative training application based on VR and mixed reality (MR) with the overlap between real and virtual chainsaws in MR. The research and training platform was developed for the Meta quest 2 head-mounted display. The research and training platform application is based on the Unity 3D engine, and Present Platform Interaction SDK (PPI-SDK) developed by Meta. PPI-SDK avoids the use of controllers and enables hand tracking and MR. With the combination of these two technologies, it was possible to overlay a virtual chainsaw with a real chainsaw in MR and synchronize their movements in VR. This ensures that the user feels the weight of the actual chainsaw, tightens the muscles, and performs the appropriate movements during the test allowing the user to learn the correct body posture. The chainsaw works only if the right sequence of cuts is made to felling the tree. Contact detection is done by Unity's physics system, which allows the interaction of objects that simulate real-world behavior. Each cut of the chainsaw is defined by a so-called collider, and the felling of the tree can only occur if the colliders are activated in the right order simulating a safe technique felling. In this way, the user can learn how to use the chainsaw safely. The system is also multiplayer, so the student and the instructor can experience VR together in a symmetrical and collaborative way. The platform simulates the following tree-felling situations with safe techniques: cutting the tree tilted forward, cutting the medium-sized tree tilted backward, cutting the large tree tilted backward, sectioning the trunk on the ground, and cutting branches. The application is being evaluated on a sample of university students through a special questionnaire. The results are expected to test both the increase in learning compared to a theoretical lecture and the immersive and telepresence of the platform.

Keywords: chainsaw, collaborative symmetric virtual reality, mixed reality, operator training

Procedia PDF Downloads 103
282 Elicitation Methods of Requirements Gathering in Shopping Mobile Application Development

Authors: Xiao Yihong, Li Zhixuan, Wong Kah Seng, Shen Xingcang

Abstract:

Requirement Elicitation is one of the important factors in developing any new application. Most systems fail just because of wrong elicitation practice. As a result, developers always choose different methods in different fields to achieve optimal results. This paper analyses four cases to understand the effectiveness of different requirement elicitation methods in the field of mobile shopping applications. The elicitation methods we studied included interviews, questionnaires, prototypes, analysis of existing systems, focus groups, brainstorming, and so on. Through the research and analysis results, we ensured the need for a mixture of elicitation methods. Meanwhile, the method adopted should be determined according to the scale of the project and be operated in a reasonable order to ensure the high efficiency of requirement elicitation.

Keywords: requirements elicitation method, shopping, mobile application, software requirement engineering

Procedia PDF Downloads 115
281 The Comparison of Open Source Software for Digital Libraries

Authors: Kanita Beširević

Abstract:

Open-source software development activities highly rely on Internet gathering communities volunteering in software development projects. Additionally, the libraries and cultural institutions share their metadata in the form of linked metadata to enable dissemination and enrichment. The open-source software provides free alternatives to traditional software solutions. The article aims to investigate the ever-increasing options for the digital library open source software adoption. The software available is presented and compared to other software solutions as well as to their previous versions. The top three open-source digital library software solutions are presented and compared. The comparison criteria are adopted from the UNESCO study by Bankier, J., & Gleason, K. Institutional Repository Software Comparison comprising of twelve criteria to appraise software, namely: infrastructure, front-end design, content discovery, publication tools, interoperability, and preservation. This article adopts a descriptive methodology based on data and information collected through selected software websites and the literature review.

Keywords: open source software, digital library, DSpace, Fedora, Greenstone

Procedia PDF Downloads 111
280 Methods Used to Perform Requirements Elicitation for Healthcare Software Development

Authors: Tang Jiacheng, Fang Tianyu, Liu Yicen, Xiang Xingzhou

Abstract:

The proportion of healthcare services is increasing throughout the globe. The convergence of mobile technology is driving new business opportunities, innovations in healthcare service delivery and the promise of a better life tomorrow for different populations with various healthcare needs. One of the most important phases for the combination of health care and mobile applications is to elicit requirements correctly. In this paper, four articles from different research directions with four topics on healthcare were detailed analyzed and summarized. We identified the underlying problems in guidance to develop mobile applications to provide healthcare service for Older adults, Women in menopause, Patients undergoing covid. These case studies cover several elicitation methods: survey, prototyping, focus group interview and questionnaire. And the effectiveness of these methods was analyzed along with the advantages and limitations of these methods, which is beneficial to adapt the elicitation methods for future software development process.

Keywords: healthcare, software requirement elicitation, mobile applications, prototyping, focus group interview

Procedia PDF Downloads 143
279 K-Means Clustering-Based Infinite Feature Selection Method

Authors: Seyyedeh Faezeh Hassani Ziabari, Sadegh Eskandari, Maziar Salahi

Abstract:

Infinite Feature Selection (IFS) algorithm is an efficient feature selection algorithm that selects a subset of features of all sizes (including infinity). In this paper, we present an improved version of it, called clustering IFS (CIFS), by clustering the dataset in advance. To do so, first, we apply the K-means algorithm to cluster the dataset, then we apply IFS. In the CIFS method, the spatial and temporal complexities are reduced compared to the IFS method. Experimental results on 6 datasets show the superiority of CIFS compared to IFS in terms of accuracy, running time, and memory consumption.

Keywords: feature selection, infinite feature selection, clustering, graph

Procedia PDF Downloads 122
278 A Method for Quantitative Assessment of the Dependencies between Input Signals and Output Indicators in Production Systems

Authors: Maciej Zaręba, Sławomir Lasota

Abstract:

Knowing the degree of dependencies between the sets of input signals and selected sets of indicators that measure a production system's effectiveness is of great importance in the industry. This paper introduces the SELM method that enables the selection of sets of input signals, which affects the most the selected subset of indicators that measures the effectiveness of a production system. For defined set of output indicators, the method quantifies the impact of input signals that are gathered in the continuous monitoring production system.

Keywords: manufacturing operation management, signal relationship, continuous monitoring, production systems

Procedia PDF Downloads 110
277 Method and Apparatus for Optimized Job Scheduling in the High-Performance Computing Cloud Environment

Authors: Subodh Kumar, Amit Varde

Abstract:

Typical on-premises high-performance computing (HPC) environments consist of a fixed number and a fixed set of computing hardware. During the design of the HPC environment, the hardware components, including but not limited to CPU, Memory, GPU, and networking, are carefully chosen from select vendors for optimal performance. High capital cost for building the environment is a prime factor influencing the design environment. A class of software called “Job Schedulers” are critical to maximizing these resources and running multiple workloads to extract the maximum value for the high capital cost. In principle, schedulers work by preventing workloads and users from monopolizing the finite hardware resources by queuing jobs in a workload. A cloud-based HPC environment does not have the limitations of fixed (type of and quantity of) hardware resources. In theory, users and workloads could spin up any number and type of hardware resource. This paper discusses the limitations of using traditional scheduling algorithms for cloud-based HPC workloads. It proposes a new set of features, called “HPC optimizers,” for maximizing the benefits of the elasticity and scalability of the cloud with the goal of cost-performance optimization of the workload.

Keywords: high performance computing, HPC, cloud computing, optimization, schedulers

Procedia PDF Downloads 83
276 A Formal Verification Approach for Linux Kernel Designing

Authors: Zi Wang, Xinlei He, Jianghua Lv, Yuqing Lan

Abstract:

Kernel though widely used, is complicated. Errors caused by some bugs are often costly. Statically, more than half of the mistakes occur in the design phase. Thus, we introduce a modeling method, KMVM (Linux Kernel Modeling and verification Method), based on type theory for proper designation and correct exploitation of the Kernel. In the model, the Kernel is separated into six levels: subsystem, dentry, file, struct, func, and base. Each level is treated as a type. The types are specified in the structure and relationship. At the same time, we use a demanding path to express the function to be implemented. The correctness of the design is verified by recursively checking the type relationship and type existence. The method has been applied to verify the OPEN business of VFS (virtual file system) in Linux Kernel. Also, we have designed and developed a set of security communication mechanisms in the Kernel with verification.

Keywords: formal approach, type theory, Linux Kernel, software program

Procedia PDF Downloads 125
275 The Comparative Study of Binary Artifact Repository Managers

Authors: Evgeny Chugunnyy, Alena Gerasimova, Kirill Chernyavskiy, Alexander Krasnov

Abstract:

One of the primary component of Continuous deployment (CD) is a binary artifact repository — the place where artifacts are stored with metadata in a structured way. The binary artifact repository manager (BARM) is a software, which implements this repository logic and exposes a public application programming interface (API) for managing these artifacts. Almost every programming language ecosystem has its own artifact repository kind. During creating Artipie — BARM constructor and server, we analyzed and implemented a lot of different artifact repositories. In this paper we present criterias for comparing artifact repositories, and analyze the most popular repositories using these metrics. We also describe some of the notable features of different repositories. This paper aimed to help people who are creating, maintaining or optimizing software repository and CI tools.

Keywords: artifact, repository, continuous deployment, build automation, artifacts management

Procedia PDF Downloads 141
274 Deep Supervision Based-Unet to Detect Buildings Changes from VHR Aerial Imagery

Authors: Shimaa Holail, Tamer Saleh, Xiongwu Xiao

Abstract:

Building change detection (BCD) from satellite imagery is an essential topic in urbanization monitoring, agricultural land management, and updating geospatial databases. Recently, methods for detecting changes based on deep learning have made significant progress and impressive results. However, it has the problem of being insensitive to changes in buildings with complex spectral differences, and the features being extracted are not discriminatory enough, resulting in incomplete buildings and irregular boundaries. To overcome these problems, we propose a dual Siamese network based on the Unet model with the addition of a deep supervision strategy (DS) in this paper. This network consists of a backbone (encoder) based on ImageNet pre-training, a fusion block, and feature pyramid networks (FPN) to enhance the step-by-step information of the changing regions and obtain a more accurate BCD map. To train the proposed method, we created a new dataset (EGY-BCD) of high-resolution and multi-temporal aerial images captured over New Cairo in Egypt to detect building changes for this purpose. The experimental results showed that the proposed method is effective and performs well with the EGY-BCD dataset regarding the overall accuracy, F1-score, and mIoU, which were 91.6 %, 80.1 %, and 73.5 %, respectively.

Keywords: building change detection, deep supervision, semantic segmentation, EGY-BCD dataset

Procedia PDF Downloads 109
273 Cloud Computing Security for Multi-Cloud Service Providers: Controls and Techniques in Our Modern Threat Landscape

Authors: Sandesh Achar

Abstract:

Cloud computing security is a broad term that covers a variety of security concerns for organizations that use cloud services. Multi-cloud service providers must consider several factors when addressing security for their customers, including identity and access management, data at rest and in transit, egress and ingress traffic control, vulnerability and threat management, and auditing. This paper explores each of these aspects of cloud security in detail and provides recommendations for best practices for multi-cloud service providers. It also discusses the challenges inherent in securing a multi-cloud environment and offers solutions for overcoming these challenges. By the end of this paper, readers should have a good understanding of the various security concerns associated with multi-cloud environments in the context of today’s modern cyber threats and how to address them.

Keywords: multi-cloud service, system organization control, data loss prevention, identity and access management

Procedia PDF Downloads 94
272 Secure Authentication Scheme Based on Numerical Series Cryptography for Internet of Things

Authors: Maha Aladdin, Khaled Nagaty, Abeer Hamdy

Abstract:

The rapid advancement cellular networks and wireless networks have laid a solid basis for the Internet of Things. IoT has evolved into a unique standard that allows diverse physical devices to collaborate with one another. A service provider gives a variety of services that may be accessed via smart apps anywhere, at any time, and from any location over the Internet. Because of the public environment of mobile communication and the Internet, these services are highly vulnerable to a several malicious attacks, such as unauthorized disclosure by hostile attackers. As a result, the best option for overcoming these vulnerabilities is a strong authentication method. In this paper, a lightweight authentication scheme that is based on numerical series cryptography is proposed for the IoT environments. It allows mutual authentication between IoT devices Parametric study and formal proofs are utilized to illustrate that the pro-posed approach is resistant to a variety of security threats.

Keywords: internet of things, authentication, cryptography, security protocol

Procedia PDF Downloads 114
271 Prioritization of Mutation Test Generation with Centrality Measure

Authors: Supachai Supmak, Yachai Limpiyakorn

Abstract:

Mutation testing can be applied for the quality assessment of test cases. Prioritization of mutation test generation has been a critical element of the industry practice that would contribute to the evaluation of test cases. The industry generally delivers the product under the condition of time to the market and thus, inevitably sacrifices software testing tasks, even though many test cases are required for software verification. This paper presents an approach of applying a social network centrality measure, PageRank, to prioritize mutation test generation. The source code with the highest values of PageRank will be focused first when developing their test cases as these modules are vulnerable to defects or anomalies which may cause the consequent defects in many other associated modules. Moreover, the approach would help identify the reducible test cases in the test suite, still maintaining the same criteria as the original number of test cases.

Keywords: software testing, mutation test, network centrality measure, test case prioritization

Procedia PDF Downloads 104
270 A Multilayer Perceptron Neural Network Model Optimized by Genetic Algorithm for Significant Wave Height Prediction

Authors: Luis C. Parra

Abstract:

The significant wave height prediction is an issue of great interest in the field of coastal activities because of the non-linear behavior of the wave height and its complexity of prediction. This study aims to present a machine learning model to forecast the significant wave height of the oceanographic wave measuring buoys anchored at Mooloolaba of the Queensland Government Data. Modeling was performed by a multilayer perceptron neural network-genetic algorithm (GA-MLP), considering Relu(x) as the activation function of the MLPNN. The GA is in charge of optimized the MLPNN hyperparameters (learning rate, hidden layers, neurons, and activation functions) and wrapper feature selection for the window width size. Results are assessed using Mean Square Error (MSE), Root Mean Square Error (RMSE), and Mean Absolute Error (MAE). The GAMLPNN algorithm was performed with a population size of thirty individuals for eight generations for the prediction optimization of 5 steps forward, obtaining a performance evaluation of 0.00104 MSE, 0.03222 RMSE, 0.02338 MAE, and 0.71163% of MAPE. The results of the analysis suggest that the MLPNNGA model is effective in predicting significant wave height in a one-step forecast with distant time windows, presenting 0.00014 MSE, 0.01180 RMSE, 0.00912 MAE, and 0.52500% of MAPE with 0.99940 of correlation factor. The GA-MLP algorithm was compared with the ARIMA forecasting model, presenting better performance criteria in all performance criteria, validating the potential of this algorithm.

Keywords: significant wave height, machine learning optimization, multilayer perceptron neural networks, evolutionary algorithms

Procedia PDF Downloads 102
269 UNIX Source Code Leak: Evaluation and Feasible Solutions

Authors: Gu Dongxing, Li Yuxuan, Nong Tengxiao, Burra Venkata Durga Kumar

Abstract:

Since computers are widely used in business models, more and more companies choose to store important information in computers to improve productivity. However, this information can be compromised in many cases, such as when it is stored locally on the company's computers or when it is transferred between servers and clients. Of these important information leaks, source code leaks are probably the most costly. Because the source code often represents the core technology of the company, especially for the Internet companies, source code leakage may even lead to the company's core products lose market competitiveness, and then lead to the bankruptcy of the company. In recent years, such as Microsoft, AMD and other large companies have occurred source code leakage events, suffered a huge loss. This reveals to us the importance and necessity of preventing source code leakage. This paper aims to find ways to prevent source code leakage based on the direction of operating system, and based on the fact that most companies use Linux or Linux-like system to realize the interconnection between server and client, to discuss how to reduce the possibility of source code leakage during data transmission.

Keywords: data transmission, Linux, source code, operating system

Procedia PDF Downloads 260
268 Time Travel Testing: A Mechanism for Improving Renewal Experience

Authors: Aritra Majumdar

Abstract:

While organizations strive to expand their new customer base, retaining existing relationships is a key aspect of improving overall profitability and also showcasing how successful an organization is in holding on to its customers. It is an experimentally proven fact that the lion’s share of profit always comes from existing customers. Hence seamless management of renewal journeys across different channels goes a long way in improving trust in the brand. From a quality assurance standpoint, time travel testing provides an approach to both business and technology teams to enhance the customer experience when they look to extend their partnership with the organization for a defined phase of time. This whitepaper will focus on key pillars of time travel testing: time travel planning, time travel data preparation, and enterprise automation. Along with that, it will call out some of the best practices and common accelerator implementation ideas which are generic across verticals like healthcare, insurance, etc. In this abstract document, a high-level snapshot of these pillars will be provided. Time Travel Planning: The first step of setting up a time travel testing roadmap is appropriate planning. Planning will include identifying the impacted systems that need to be time traveled backward or forward depending on the business requirement, aligning time travel with other releases, frequency of time travel testing, preparedness for handling renewal issues in production after time travel testing is done and most importantly planning for test automation testing during time travel testing. Time Travel Data Preparation: One of the most complex areas in time travel testing is test data coverage. Aligning test data to cover required customer segments and narrowing it down to multiple offer sequencing based on defined parameters are keys for successful time travel testing. Another aspect is the availability of sufficient data for similar combinations to support activities like defect retesting, regression testing, post-production testing (if required), etc. This section will talk about the necessary steps for suitable data coverage and sufficient data availability from a time travel testing perspective. Enterprise Automation: Time travel testing is never restricted to a single application. The workflow needs to be validated in the downstream applications to ensure consistency across the board. Along with that, the correctness of offers across different digital channels needs to be checked in order to ensure a smooth customer experience. This section will talk about the focus areas of enterprise automation and how automation testing can be leveraged to improve the overall quality without compromising on the project schedule. Along with the above-mentioned items, the white paper will elaborate on the best practices that need to be followed during time travel testing and some ideas pertaining to accelerator implementation. To sum it up, this paper will be written based on the real-time experience author had on time travel testing. While actual customer names and program-related details will not be disclosed, the paper will highlight the key learnings which will help other teams to implement time travel testing successfully.

Keywords: time travel planning, time travel data preparation, enterprise automation, best practices, accelerator implementation ideas

Procedia PDF Downloads 153
267 Stack Overflow Detection and Prevention on Operating Systems Using Machine Learning and Control-Flow Enforcement Technology

Authors: Cao Jiayu, Lan Ximing, Huang Jingjia, Burra Venkata Durga Kumar

Abstract:

The first virus to attack personal computers was born in early 1986, called C-Brain, written by a pair of Pakistani brothers. In those days, people still used dos systems, manipulating computers with the most basic command lines. In the 21st century today, computer performance has grown geometrically. But computer viruses are also evolving and escalating. We never stop fighting against security problems. Stack overflow is one of the most common security vulnerabilities in operating systems. It may result in serious security issues for an operating system if a program in it has a vulnerability with administrator privileges. Certain viruses change the value of specific memory through a stack overflow, allowing computers to run harmful programs. This study developed a mechanism to detect and respond to time whenever a stack overflow occurs. We demonstrate the effectiveness of standard machine learning algorithms and control flow enforcement techniques in predicting computer OS security using generating suspicious vulnerability functions (SVFS) and associated suspect areas (SAS). The method can minimize the possibility of stack overflow attacks occurring.

Keywords: operating system, security, stack overflow, buffer overflow, machine learning, control-flow enforcement technology

Procedia PDF Downloads 111
266 A Study on Unix Process Crash Based on Efficient Process Management Method

Authors: Guo Haonan, Chen Peiyu, Zhao Hanyu, Burra Venkata Durga Kumar

Abstract:

Unix and Unix-like operating systems are widely used due to their high stability but are limited by the parent-child process structure, and the child process depends on the parent process, so the crash of a single process may cause the entire process group or even the entire system to fail. Another possibility of unexpected process termination is that the system administrator inadvertently closed the terminal or pseudo-terminal where the application was launched, causing the application process to terminate unexpectedly. This paper mainly analyzes the reasons for the problems and proposes two solutions.

Keywords: process management, daemon, login-bash and non-login bash, process group

Procedia PDF Downloads 129