Search results for: mathematical algorithms of targeting and persecution

Authors: Annette Cosgrove

Abstract:

The objectives of this research study is to examine the impact of professional development in Technology Enhanced Learning (TEL) and the digitisation of learning in teaching communities across multiple higher education sites in the ATU (Atlantic Technological University *) ( 2020-2025), including the proposal of an evidence based digital teaching model for use in a future pandemic. The research strategy undertaken for this PhD Study is a multi-site study using mixed methods. Qualitative & quantitative methods are being used in the study to collect data. A pilot study was carried out initially , feedback collected and the research instrument was edited to reflect this feedback, before being administered. The purpose of the staff questionnaire is to evaluate the impact of professional development in the area of TEL, and to capture the practitioners views on the perceived impact on their teaching practice in the higher education sector across ATU (West of Ireland – 5 Higher education locations ). The phenomenon being explored is ‘ the impact of professional development in the area of technology enhanced learning and on teaching practice in a higher education institution.’ The research methodology chosen for this study is an Action based Research Study. The researcher has chosen this approach as it is a prime strategy for developing educational theory and enhancing educational practice . This study includes quantitative and qualitative methods to elicit data which will quantify the impact that continuous professional development in the area of digital teaching practice and technologies has on the practitioner’s teaching practice in higher education. The research instruments / data collection tools for this study include a lecturer survey with a targeted TEL Practice group ( Pre and post covid experience) and semi-structured interviews with lecturers.. This research is currently being conducted across the ATU multisite campus and targeting Higher education lecturers that have completed formal CPD in the area of digital teaching. ATU, a west of Ireland university is the focus of the study , The research questionnaire has been deployed, with 75 respondents to date across the ATU - the primary questionnaire and semi- formal interviews are ongoing currently – the purpose being to evaluate the impact of formal professional development in the area of TEL and its perceived impact on the practitioners teaching practice in the area of digital teaching and learning . This paper will present initial findings, reflections and data from this ongoing research study.

Keywords: TEL, DTL, digital teaching, digital assessment

Procedia PDF Downloads 50

Authors: Mo-Se Lee, Cheol-Hwi Ahn, Kee-Young Kwahk, Hyunchul Ahn

Abstract:

Over the past decade, deep learning has been in spotlight among various machine learning algorithms. In particular, CNN (Convolutional Neural Network), which is known as effective solution for recognizing and classifying images, has been popularly applied to classification and prediction problems in various fields. In this study, we try to apply CNN to stock market prediction, one of the most challenging tasks in the machine learning research. In specific, we propose to apply CNN as the binary classifier that predicts stock market direction (up or down) by using a graph as its input. That is, our proposal is to build a machine learning algorithm that mimics a person who looks at the graph and predicts whether the trend will go up or down. Our proposed model consists of four steps. In the first step, it divides the dataset into 5 days, 10 days, 15 days, and 20 days. And then, it creates graphs for each interval in step 2. In the next step, CNN classifiers are trained using the graphs generated in the previous step. In step 4, it optimizes the hyper parameters of the trained model by using the validation dataset. To validate our model, we will apply it to the prediction of KOSPI200 for 1,986 days in eight years (from 2009 to 2016). The experimental dataset will include 14 technical indicators such as CCI, Momentum, ROC and daily closing price of KOSPI200 of Korean stock market.

Keywords: convolutional neural network, deep learning, Korean stock market, stock market prediction

Procedia PDF Downloads 413

Authors: Gideon C. Joseph

Abstract:

The ability to localize mobile devices is quite important, as some applications may require location information of these devices to operate or deliver better services to the users. Although there are several ways of acquiring location data of mobile devices, the WLAN fingerprinting approach has been considered in this work. This approach uses the Received Signal Strength Indicator (RSSI) measurement as a function of the position of the mobile device. RSSI is a quantitative technique of describing the radio frequency power carried by a signal. RSSI may be used to determine RF link quality and is very useful in dense traffic scenarios where interference is of major concern, for example, indoor environments. This research aims to design a system that can predict the location of a mobile device, when supplied with the mobile’s RSSIs. The developed system takes as input the RSSIs relating to the mobile device, and outputs parameters that describe the location of the device such as the longitude, latitude, floor, and building. The relationship between the Received Signal Strengths (RSSs) of mobile devices and their corresponding locations is meant to be modelled; hence, subsequent locations of mobile devices can be predicted using the developed model. It is obvious that describing mathematical relationships between the RSSIs measurements and localization parameters is one option to modelling the problem, but the complexity of such an approach is a serious turn-off. In contrast, we propose an intelligent system that can learn the mapping of such RSSIs measurements to the localization parameters to be predicted. The system is capable of upgrading its performance as more experiential knowledge is acquired. The most appealing consideration to using such a system for this task is that complicated mathematical analysis and theoretical frameworks are excluded or not needed; the intelligent system on its own learns the underlying relationship in the supplied data (RSSI levels) that corresponds to the localization parameters. These localization parameters to be predicted are of two different tasks: Longitude and latitude of mobile devices are real values (regression problem), while the floor and building of the mobile devices are of integer values or categorical (classification problem). This research work presents artificial neural network based intelligent systems to model the relationship between the RSSIs predictors and the mobile device localization parameters. The designed systems were trained and validated on the collected WLAN fingerprint database. The trained networks were then tested with another supplied database to obtain the performance of trained systems on achieved Mean Absolute Error (MAE) and error rates for the regression and classification tasks involved therein.

Keywords: indoor localization, WLAN fingerprinting, neural networks, classification, regression

Procedia PDF Downloads 327

Authors: Cheng-Jhe Lee, Chiun-Chieh Hsu

Abstract:

Due to the rapid development of the Internet and the increased availability of digital documents, the excessive information on the Internet has led to information overflow problem. In order to solve these problems for effective information retrieval, document clustering in text mining becomes a popular research topic. Clustering is the unsupervised classification of data items into groups without the need of training data. Many conventional document clustering methods perform inefficiently for large document collections because they were originally designed for relational database. Therefore they are impractical in real-world document clustering and require special handling for high dimensionality and high volume. We propose the FIHC (Frequent Itemset-based Hierarchical Clustering) method, which is a hierarchical clustering method developed for document clustering, where the intuition of FIHC is that there exist some common words for each cluster. FIHC uses such words to cluster documents and builds hierarchical topic tree. In this paper, we combine FIHC algorithm with ontology to solve the semantic problem and mine the meaning behind the words in documents. Furthermore, we use the closed frequent itemsets instead of only use frequent itemsets, which increases efficiency and scalability. The experimental results show that our method is more accurate than those of well-known document clustering algorithms.

Keywords: FIHC, documents clustering, ontology, closed frequent itemset

Procedia PDF Downloads 377

Authors: Ben Clapperton, Daniel Stahl, Kimberley Goldsmith, Trudie Chalder

Abstract:

Chronic fatigue syndrome (CFS) is a condition characterised by chronic disabling fatigue and other symptoms that currently can't be explained by any underlying medical condition. Although clinical trials support the effectiveness of cognitive behaviour therapy (CBT), the success rate for individual patients is modest. Patients vary in their response and little is known which factors predict or moderate treatment outcomes. The aim of the project is to develop a prediction model from baseline characteristics of patients, such as demographics, clinical and psychological variables, which may predict likely treatment outcome and provide guidance for clinical decision making and help clinicians to recommend the best treatment. The project is aimed at identifying subgroups of patients with similar baseline characteristics that are predictive of treatment effects using modern cluster analyses and data mining machine learning algorithms. The characteristics of these groups will then be used to inform the types of individuals who benefit from a specific treatment. In addition, results will provide a better understanding of for whom the treatment works. The suitability of different clustering methods to identify subgroups and their response to different treatments of CFS patients is compared.

Keywords: chronic fatigue syndrome, latent class analysis, prediction modelling, self-organizing maps

Procedia PDF Downloads 211

Authors: Mohamed K. Metwally, Sherif H. El-Gohary, Kyung Min Byun, Seung Moo Han, Soo Yeol Lee, Min Hyoung Cho, Gon Khang, Jinsung Cho, Tae-Seong Kim

Abstract:

Photoacoustic imaging (PAI) is a non-invasive and non-ionizing imaging modality that combines the absorption contrast of light with ultrasound resolution. Laser is used to deposit optical energy into a target (i.e., optical fluence). Consequently, the target temperature rises, and then thermal expansion occurs that leads to generating a PA signal. In general, most image reconstruction algorithms for PAI assume uniform fluence within an imaging object. However, it is known that optical fluence distribution within the object is non-uniform. This could affect the reconstruction of PA images. In this study, we have investigated the influence of optical fluence distribution on PA back-propagation imaging using finite element method. The uniform fluence was simulated as a triangular waveform within the object of interest. The non-uniform fluence distribution was estimated by solving light propagation within a tissue model via Monte Carlo method. The results show that the PA signal in the case of non-uniform fluence is wider than the uniform case by 23%. The frequency spectrum of the PA signal due to the non-uniform fluence has missed some high frequency components in comparison to the uniform case. Consequently, the reconstructed image with the non-uniform fluence exhibits a strong smoothing effect.

Keywords: finite element method, fluence distribution, Monte Carlo method, photoacoustic imaging

Procedia PDF Downloads 367

Authors: Marek Dlapa

Abstract:

The algebraic approach is applied to the control of the HiMAT (Highly Maneuverable Aircraft Technology). The objective is to find a robust controller which guarantees robust stability and decoupled control of longitudinal model of a scaled remotely controlled vehicle version of the advanced fighter HiMAT. Control design is performed by decoupling the nominal MIMO (multi-input multi-output) system into two identical SISO (single-input single-output) plants which are approximated by a 4th order transfer function. The algebraic approach is then used for pole placement design, and the nominal closed-loop poles are tuned so that the peak of the µ-function is minimal. As an optimization tool, evolutionary algorithm Differential Migration is used in order to overcome the multimodality of the cost function yielding simple controller with decoupling for nominal plant which is compared with the D-K iteration through simulations of standard longitudinal manoeuvres documenting decoupled control obtained from algebraic approach for nominal plant as well as worst case perturbation.

Keywords: algebraic approach, evolutionary computation, genetic algorithms, HiMAT, robust control, structured singular value

Procedia PDF Downloads 126

Authors: Martin Kendra, Jaroslav Mašek, Juraj Čamaj, Martin Búda

Abstract:

The paper deals with possibilities of increase train capacity by using a new type of railway wagon. In the first part is created a mathematical model to calculate the capacity of the train. The model is based on the main limiting parameters of the train - maximum number of axles per train, the maximum gross weight of the train, the maximum length of train and number of TEUs per one wagon. In the second part is the model applied to four different model trains with different composition of the train set and three different average weights of TEU and a train consisting of a new type of wagons. The result is to identify where the carrying capacity of the original trains is higher, respectively less than a capacity of the train consisting of a new type of wagons.

Keywords: loading units, theoretical capacity model, train capacity, wagon for intermodal transport

Procedia PDF Downloads 479

Authors: Violeta Damjanovic-Behrendt

Abstract:

This paper presents an approach for optimal cyber security decisions to protect instances of a federated Internet of Things (IoT) platform in the cloud. The presented solution implements the repeated Stackelberg Security Game (SSG) and a model called Stochastic Human behaviour model with AttRactiveness and Probability weighting (SHARP). SHARP employs the Subjective Utility Quantal Response (SUQR) for formulating a subjective utility function, which is based on the evaluations of alternative solutions during decision-making. We augment the repeated SSG (including SHARP and SUQR) with a reinforced learning algorithm called Naïve Q-Learning. Naïve Q-Learning belongs to the category of active and model-free Machine Learning (ML) techniques in which the agent (either the defender or the attacker) attempts to find an optimal security solution. In this way, we combine GT and ML algorithms for discovering optimal cyber security policies. The proposed security optimization components will be validated in a collaborative cloud platform that is based on the Industrial Internet Reference Architecture (IIRA) and its recently published security model.

Keywords: security, internet of things, cloud computing, stackelberg game, machine learning, naive q-learning

Procedia PDF Downloads 336

Authors: Samuel Holder

Abstract:

Mass incarceration in the United States is a human rights issue, not merely a civil rights problem. It is a human rights problem not only because the United States has a high rate of incarceration, but more importantly because of who is jailed, for what purpose they are jailed and, ultimately, the manner in which they are jailed. To sustain the scale of the criminal justice system, one of the darker policies involves a multi-tiered strategy of fee- and fine-collection, targeting, usually, the most vulnerable and poor, many of whom run into the law via small offenses that do not rise to the level of felonies. This paper advances the notion that this debt collection-to-incarceration pipeline is tantamount to a modern-day debtors’ prison system. This article seeks to confront the thorny issue of incarceration via criminal justice debt from a human rights and cause-lawyering position. It will argue that a two-pronged cause-lawyering strategy: the first focused on traditional litigation along constitutional grounds, and the second, an advocacy approach rooted in grassroots campaigns, designed to shift the normative operation and understanding of the rights of marginalized and racialized offenders. Ultimately, the argument suggests that this approach will be effective in combatting the (often highly privatized) criminal justice debt system and bring the roles of 'incapacitation, rehabilitation, deterrence, and retribution' back into the criminal justice legal conversation. Part I contextualizes and historicizes the role of fees, penalties, and fines in American criminal justice. Part II examines the emergence of private industry in the criminal justice system, and its role in the acceleration of profit-driven criminal justice debt collection and incarceration. Part III addresses the failures of the federal and state law and legislation in combatting predatory incarceration and debt collection in the criminal justice system, particularly as waged against the indigent and/or ethnically or racially marginalized. Part IV examines the potential for traditional cause-lawyering litigation along constitutional grounds, using case studies across contexts for illustration. Finally, Part V will review the radical cause-lawyer’s role in the normative struggle in redefining prisoners’ rights and the rights of the marginalized (and racialized) as they intersect at the crossroads of criminal justice debt. This paper will conclude with recommendations for litigation and advocacy, drawing on hypotheses advanced, and informed by case studies from a variety of both national and international jurisdictions.

Keywords: cause-lawyering, criminal justice debt, human rights, judicial fees

Procedia PDF Downloads 152

Authors: Ramon Erasmo Verdin Rodriguez

Abstract:

Many has been the ways that historically the managers and gurus has been trying to get closer to the perfect supply chain, but since this topic is so vast and very complex the bigger the companies are, the duty has not been certainly easy. On this research, you are going to see thru the entrails of the logistics that happens at a glass manufacturing company with the number one raw material of the process that is the silica sand. After a very quick passage thru the supply chain, this document is going to focus on the way that raw materials flow thru the system, so after that, an analysis and research can take place to improve the logistics. Thru Operations Research techniques, it will be analyzed the current scheme of distribution and inventories of raw materials at a glass company’s plants, so after a mathematical conceptualization process, the supply chain could be optimized with the purpose of reducing the uncertainty of supply and obtaining an economic benefit at the very end of this research.

Keywords: inventory management, operations research, optimization, supply chain

Procedia PDF Downloads 312

Authors: Awatef Chniguir, Tarek Farah, Zouhair Ben Jemaa, Safya Belguith

Abstract:

Optimal routing allows minimizing energy consumption in wireless sensor networks (WSN). Clustering has proven its effectiveness in organizing WSN by reducing channel contention and packet collision and enhancing network throughput under heavy load. Therefore, nowadays, with the emergence of the Internet of Things, heterogeneity is essential. Stable election protocol (SEP) that has increased the network stability period and lifetime is the first clustering protocol for heterogeneous WSN. SEP and its descendants, namely SEP, Threshold Sensitive SEP (TSEP), Enhanced TSEP (ETSSEP) and Current Energy Allotted TSEP (CEATSEP), were studied. These algorithms’ performance was evaluated based on different metrics, especially first node death (FND), to compare their stability. Simulations were conducted on the MATLAB tool considering two scenarios: The first one demonstrates the fraction variation of advanced nodes by setting the number of total nodes. The second considers the interpretation of the number of nodes while keeping the number of advanced nodes permanent. CEATSEP outperforms its antecedents by increasing stability and, at the same time, keeping a low throughput. It also operates very well in a large-scale network. Consequently, CEATSEP has a useful lifespan and energy efficiency compared to the other routing protocol for heterogeneous WSN.

Keywords: clustering, heterogeneous, stability, scalability, IoT, WSN

Procedia PDF Downloads 113

Authors: Odette Umugiraneza, Etienne Nzaramyimana

Abstract:

Confidence and beliefs are the main contributors to the improvement of teachers’ mathematical knowledge. The objective of this study was to investigate teachers’ confidence and beliefs towards technology use in teaching mathematics subjects in the Musanze District. The data were collected using closed and open questions. These were distributed to 118 secondary school senior 1 to 6 mathematics teachers in Musanze district. The findings revealed that the teachers’ confidence about the use of technology in teaching mathematics needs improvement. Apart from confidence, almost a third of the teachers convoyed negative beliefs that technology plays great importance in promoting the understanding of mathematics. Teachers as knowledge transmitters are required to join various professional courses towards technology integration in the teaching of mathematics, to improve the effectiveness of teaching and learning.

Keywords: knowledge, technology, teachers’ confidence, beliefs, barriers of technology use

Procedia PDF Downloads 109

Authors: Selisha A. Sooklal, Phelelani Mpangase, Shaun Aron, Karl Rumbold

Abstract:

Organically bound fluorine (C-F bond) is extremely rare in nature. Despite this, the first fluorinated secondary metabolite, fluoroacetate, was isolated from the plant Dichapetalum cymosum (commonly known as Gifblaar). However, the enzyme responsible for fluorination (fluorinase) in Gifblaar was never isolated and very little progress has been achieved in understanding this process in higher plants. Fluorinated compounds have vast applications in the pharmaceutical, agrochemical and fine chemicals industries. Consequently, an enzyme capable of catalysing a C-F bond has great potential as a biocatalyst in the industry considering that the field of fluorination is virtually synthetic. As with any biocatalyst, a range of these enzymes are required. Therefore, it is imperative to expand the exploration for novel fluorinases. This study aimed to gain molecular insights into secondary metabolite biosynthesis in Gifblaar using a high-throughput sequencing-based approach. Mechanical wounding studies were performed using Gifblaar leaf tissue in order to induce expression of the fluorinase. The transcriptome of the wounded and unwounded plant was then sequenced on the Illumina HiSeq platform. A total of 26.4 million short sequence reads were assembled into 77 845 transcripts using Trinity. Overall, 68.6 % of transcripts were annotated with gene identities using public databases (SwissProt, TrEMBL, GO, COG, Pfam, EC) with an E-value threshold of 1E-05. Sequences exhibited the greatest homology to the model plant, Arabidopsis thaliana (27 %). A total of 244 annotated transcripts were found to be differentially expressed between the wounded and unwounded plant. In addition, secondary metabolic pathways present in Gifblaar were successfully reconstructed using Pathway tools. Due to lack of genetic information for plant fluorinases, a transcript failed to be annotated as a fluorinating enzyme. Thus, a local database containing the 5 existing bacterial fluorinases was created. Fifteen transcripts having homology to partial regions of existing fluorinases were found. In efforts to obtain the full coding sequence of the Gifblaar fluorinase, primers were designed targeting the regions of homology and genome walking will be performed to amplify the unknown regions. This is the first genetic data available for Gifblaar. It has provided novel insights into the mechanisms of metabolite biosynthesis and will allow for the discovery of the first eukaryotic fluorinase.

Keywords: biocatalyst, fluorinase, gifblaar, transcriptome

Procedia PDF Downloads 257

Authors: Bandar Ali Al-Asbahi, Mohammad Hafizuddin Haji Jumali

Abstract:

Forster energy transfer between poly (9,9'-di-n-octylfluorenyl-2,7-diyl) (PFO)/TiO2 nanoparticles (NPs) as a donor and Fluorol 7GA as an acceptor has been studied. The energy transfer parameters were calculated by using mathematical models. The dominant mechanism responsible for the energy transfer between the donor and acceptor molecules was Forster-type, as evidenced by large values of quenching rate constant, energy transfer rate constant and critical distance of energy transfer. Moreover, these composites which were used as an emissive layer in organic light emitting diodes, were investigated in terms of current density–voltage and electroluminescence spectra.

Keywords: energy transfer parameters, forster-type, electroluminescence, organic light emitting diodes

Procedia PDF Downloads 409

Authors: Shenpu Liu

Abstract:

Space syntax, a paradigm of the urban research, which manifests people’s intuitive and abstract perception of a material space with a solid mathematical way, explores how space represents its social characteristics. Taking Xiazhuang village and Shijia Village in Huzhou as an example and focusing on inward structure and street space, this article recognizes the connotative significance of the settlement with the aid of space syntax theory and quantitative analysis method from the perspective of spatial configuration to present relevant suggestions for its future planning and provides references for traditional rural settlement protection.

Keywords: Shijia village, space configuration, space syntax, traditional rural settlement, Xiazhuang village

Procedia PDF Downloads 216

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 103

Authors: Farahnaz Sadat Golestan Hashemi, Mohd Razi Ismail, Mohd Y. Rafii

Abstract:

Clustered regularly interspaced short palindromic repeats (CRISPR) and CRISPR-associated protein (Cas) system can facilitate targeted genome editing in organisms. Dual or single guide RNA (gRNA) can program the Cas9 nuclease to cut target DNA in particular areas; thus, introducing concise mutations either via error-prone non-homologous end-joining repairing or via incorporating foreign DNAs by homologous recombination between donor DNA and target area. In spite of high demand of such promising technology, developing a well-organized procedure in order for reliable mining of potential target sites for gRNAs in large genomic data is still challenging. Hence, we aimed to perform high-throughput detection of target sites by specific PAMs for not only common Streptococcus pyogenes (SpCas9) but also for Neisseria meningitides (NmCas9) CRISPR-Cas systems. Previous research confirmed the successful application of such RNA-guided Cas9 orthologs for effective gene targeting and subsequently genome manipulation. However, Cas9 orthologs need their particular PAM sequence for DNA cleavage activity. Activity levels are based on the sequence of the protospacer and specific combinations of favorable PAM bases. Therefore, based on the specific length and sequence of PAM followed by a constant length of the target site for the two orthogonals of Cas9 protein, we created a reliable procedure to explore possible gRNA sequences. To mine CRISPR target sites, four different searching modes of sgRNA binding to target DNA strand were applied. These searching modes are as follows i) coding strand searching, ii) anti-coding strand searching, iii) both strand searching, and iv) paired-gRNA searching. Finally, a complete list of all potential gRNAs along with their locations, strands, and PAMs sequence orientation can be provided for both SpCas9 as well as another potential Cas9 ortholog (NmCas9). The artificial design of potential gRNAs in a genome of interest can accelerate functional genomic studies. Consequently, the application of such novel genome editing tool (CRISPR/Cas technology) will enhance by presenting increased versatility and efficiency.

Keywords: CRISPR/Cas9 genome editing, gRNA mining, SpCas9, NmCas9

Procedia PDF Downloads 243

Authors: Kritiyaporn Kunsook

Abstract:

Machine learning algorithms (MLAs) such us artificial neural networks (ANNs), decision tree, support vector machines (SVMs), Naïve Bayes, and ensemble classifier by voting are powerful data driven methods that are relatively less widely used in the mapping of technique of system, and thus have not been comparatively evaluated together thoroughly in this field. The performances of a series of MLAs, ANNs, decision tree, SVMs, Naïve Bayes, and ensemble classifier by voting in technique of hydroponic systems prospectively modeling are compared based on the accuracy of each model. Classification of hydroponic systems only covers the test samples from vegetables grown with Nutrient film technique (NFT) and Deep flow technique (DFT). The feature, which are the characteristics of vegetables compose harvesting height width, temperature, require light and color. The results indicate that the classification performance of the ANNs is 98%, decision tree is 98%, SVMs is 97.33%, Naïve Bayes is 96.67%, and ensemble classifier by voting is 98.96% algorithm respectively.

Keywords: artificial neural networks, decision tree, support vector machines, naïve Bayes, ensemble classifier by voting

Procedia PDF Downloads 347

Authors: Gautam Kumar Saharia, Sagardeep Talukdar, Riki Dutta, Sudipta Nandy

Abstract:

The physics informed neural network (PINN) method opens up an approach for numerically solving nonlinear partial differential equations leveraging fast calculating speed and high precession of modern computing systems. We construct the PINN based on strong universal approximation theorem and apply the initial-boundary value data and residual collocation points to weekly impose initial and boundary condition to the neural network and choose the optimization algorithms adaptive moment estimation (ADAM) and Limited-memory Broyden-Fletcher-Golfard-Shanno (L-BFGS) algorithm to optimize learnable parameter of the neural network. Next, we improve the PINN with a weighted loss function to obtain both the bright and dark soliton solutions of Fokas-Lenells equation (FLE). We find the proposed scheme of adjustable weight coefficients into PINN has a better convergence rate and generalizability than the basic PINN algorithm. We believe that the PINN approach to solve the partial differential equation appearing in nonlinear optics would be useful to study various optical phenomena.

Keywords: deep learning, optical Soliton, neural network, partial differential equation

Procedia PDF Downloads 105

Authors: Emna Benmohamed, Hela Ltifi, Mounir Ben Ayed

Abstract:

Bayesian Network (BN) is one of the most efficient classification methods. It is widely used in several fields (i.e., medical diagnostics, risk analysis, bioinformatics research). The BN is defined as a probabilistic graphical model that represents a formalism for reasoning under uncertainty. This classification method has a high-performance rate in the extraction of new knowledge from data. The construction of this model consists of two phases for structure learning and parameter learning. For solving this problem, the K2 algorithm is one of the representative data-driven algorithms, which is based on score and search approach. In addition, the integration of the expert's knowledge in the structure learning process allows the obtainment of the highest accuracy. In this paper, we propose a hybrid approach combining the improvement of the K2 algorithm called K2 algorithm for Parents and Children search (K2PC) and the expert-driven method for learning the structure of BN. The evaluation of the experimental results, using the well-known benchmarks, proves that our K2PC algorithm has better performance in terms of correct structure detection. The real application of our model shows its efficiency in the analysis of the phosphate laundry effluents' impact on the watershed in the Gafsa area (southwestern Tunisia).

Keywords: Bayesian network, classification, expert knowledge, structure learning, surface water analysis

Procedia PDF Downloads 109

Authors: Maisyatus Suadaa Irfana, Nove Eka Variant Anna, Dyah Puspitasari Sri Rahayu

Abstract:

Technological developments provide convenience to all the people. Nowadays, the communication of human with the computer is done via text. With the development of technology, human and computer communications have been conducted with a voice like communication between human beings. It provides an easy facility for many people, especially those who have special needs. Voice search technology is applied in the search of book collections in the OPAC (Online Public Access Catalog), so library visitors will find it faster and easier to find books that they need. Integration with Google is needed to convert the voice into text. To optimize the time and the results of searching, Server will download all the book data that is available in the server database. Then, the data will be converted into JSON format. In addition, the incorporation of some algorithms is conducted including Decomposition (parse) in the form of array of JSON format, the index making, analyzer to the result. It aims to make the process of searching much faster than the usual searching in OPAC because the data are directly taken to the database for every search warrant. Data Update Menu is provided with the purpose to enable users perform their own data updates and get the latest data information.

Keywords: OPAC, voice, searching, faster

Procedia PDF Downloads 326

Authors: L. K. Davis

Abstract:

The effects of the evolution force are observable in nature at all structural levels ranging from small molecular systems to conversely enormous biospheric systems. However, the evolution force and work associated with formation of biological structures has yet to be described mathematically or theoretically. In addressing the conundrum, we consider evolution from a unique perspective and in doing so we introduce the “Fundamental Theory of the Evolution Force: FTEF”. We utilized synthetic evolution artificial intelligence (SYN-AI) to identify genomic building blocks and to engineer 14-3-3 ζ docking proteins by transforming gene sequences into time-based DNA codes derived from protein hierarchical structural levels. The aforementioned served as templates for random DNA hybridizations and genetic assembly. The application of hierarchical DNA codes allowed us to fast forward evolution, while dampening the effect of point mutations. Natural selection was performed at each hierarchical structural level and mutations screened using Blosum 80 mutation frequency-based algorithms. Notably, SYN-AI engineered a set of three architecturally conserved docking proteins that retained motion and vibrational dynamics of native Bos taurus 14-3-3 ζ.

Keywords: 14-3-3 docking genes, synthetic protein design, time-based DNA codes, writing DNA code from scratch

Procedia PDF Downloads 94

Authors: Dongkyu Lee, Thanh Banh Thien, Soomi Shin

Abstract:

In the present study, the isogeometric topology optimization is proposed for cracked structures through using Solid Isotropic Material with Penalization (SIMP) as a design model. Design density variables defined in the variable space are used to approximate the element analysis density by the bivariate B-spline basis functions. The mathematical formulation of topology optimization problem solving minimum structural compliance is an alternating active-phase algorithm with the Gauss-Seidel version as an optimization model of optimality criteria. Stiffness and adjoint sensitivity formulations linked to strain energy of cracked structure are proposed in terms of design density variables. Numerical examples demonstrate interactions of topology optimization to structures design with cracks.

Keywords: topology optimization, isogeometric, NURBS, design

Procedia PDF Downloads 474

Authors: Jinming Du

Abstract:

With the COVID-19 pandemic beginning in 2020, many higher education institutions and education systems are turning to hybrid or fully distance online courses to maintain social distance and provide a safe virtual space for learning and teaching. However, the majority of faculty members were not well prepared for the shift to blended or distance learning. Communication frustrations are prevalent in both hybrid and full-distance courses. A systematic literature review was conducted by a comprehensive analysis of 1688 publications that focused on the application of the adoption of chatbots in education. This study aimed to explore instructors' experiences with chatbots in online and blended undergraduate English courses. Language learners are overwhelmed by the variety of information offered by many online sites. The recently emerged chatbots (e.g.: ChatGPT) are slightly superior in performance as compared to those traditional through previous technologies such as tapes, video recorders, and websites. The field of chatbots has been intensively researched, and new methods have been developed to demonstrate how students can best learn and practice a new language in the target language. However, it is believed that among the many areas where chatbots are applied, while chatbots have been used as effective tools for communicating with business customers, in consulting and targeting areas, and in the medical field, chatbots have not yet been fully explored and implemented in the field of language education. This issue is challenging enough for language teachers; they need to study and conduct research carefully to clarify it. Pedagogical chatbots may alleviate the perception of a lack of communication and feedback from instructors by interacting naturally with students through scaffolding the understanding of those learners, much like educators do. However, educators and instructors lack the proficiency to effectively operate this emerging AI chatbot technology and require comprehensive study or structured training to attain competence. There is a gap between language teachers’ perceptions and recent advances in the application of AI chatbots to language learning. The results of the study found that although the teachers felt that the chatbots did the best job of giving feedback, the teachers needed additional training to be able to give better instructions and to help them assist in teaching. Teachers generally perceive the utilization of chatbots to offer substantial assistance to English language instruction.

Keywords: artificial intelligence in education, chatbots, education and technology, education system, pedagogical chatbot, chatbots and language education

Procedia PDF Downloads 44

Authors: K. Djellabi, M. E. H. Latreche

Abstract:

Induction heating computer simulation is a powerful tool for process design and optimization, induction coil design, equipment selection, as well as education and business presentations. The authors share their vast experience in the practical use of computer simulation for different induction heating and heat treating processes. In this paper deals with mathematical modeling and numerical simulation of induction heating furnaces with axisymmetric geometries. For the numerical solution, we propose finite element methods combined with boundary (FEM) for the electromagnetic model using COMSOL® Multiphysics Software. Some numerical results for an industrial furnace are shown with high frequency.

Keywords: numerical methods, induction furnaces, induction heating, finite element method, Comsol multiphysics software

Procedia PDF Downloads 430

Authors: Gerard Van Es

Abstract:

In this study, the use of a high resolution surface laser scanner to assess the micro texture of runway surfaces was investigated experimentally. Micro texture is one of the important surface components that helps to provide high braking friction between aircraft tires and a wet runway surface. Algorithms to derive different parameters that characterise micro texture was developed. Surface scans with a high resolution laser scanner were conducted on 40 different runway (like) surfaces. For each surface micro texture parameters were calculated from the laser scan data. These results were correlated with results obtained from a British pendulum tester that was used on the same surface. Results obtained with the British pendulum tester are generally considered to be indicative for the micro texture related friction characteristics. The results show that a meaningful correlation can be found between different parameters that characterise micro texture obtained with the laser scanner and the British pendulum tester results. Surface laser scanners are easier to operate and give more consistent results than a British pendulum tester. Therefore for airport operators surface laser scanners can be a useful tool to determine if their runway becomes slippery when wet due to a smooth micro texture.

Keywords: runway friction, micro texture, aircraft braking performance, slippery runways

Procedia PDF Downloads 94

Authors: Kunal Tongaria, Abhishek Mangal, S. Mandal, Devendra Mohan

Abstract:

The carbonation of concrete is a phenomenon which is a function of various interdependent parameters. Therefore, in spite of numerous literature and database, the useful generalization is not an easy task. These interdependent parameters can be grouped under the category of internal and external factors. This paper focuses on the internal parameters which govern and increase the probability of the ingress of deleterious substances into concrete. The mechanism of effects of internal parameters such as microstructure for with and without supplementary cementing materials (SCM), water/binder ratio, the age of concrete etc. has been discussed. This is followed by the comparison of various proposed mathematical models for the deterioration of concrete. Based on existing laboratory experiments as well as field results, this paper concludes the present understanding of mechanism, modeling and future research needs in this field.

Keywords: carbonation, diffusion coefficient, microstructure of concrete, reinforced concrete

Procedia PDF Downloads 392

Authors: Mohamed Raouf Benmakrelouf, Joseph Rynkiewicz

Abstract:

Targeted maximum likelihood estimation (TMLE) is well-established method for causal effect estimation with desirable statistical properties. TMLE is a doubly robust maximum likelihood based approach that includes a secondary targeting step that optimizes the target statistical parameter. A causal interpretation of the statistical parameter requires assumptions of the Rubin causal framework. The causal effect of binary variable, E, on outcomes, Y, is defined in terms of comparisons between two potential outcomes as E[YE=1 − YE=0]. Our aim in this paper is to present an adaptation of TMLE methodology to estimate the causal effect of a non-binary categorical variable, providing a large application. We propose coding on the initial data in order to operate a binarization of the interest variable. For each category, we get a transformation of the non-binary interest variable into a binary variable, taking value 1 to indicate the presence of category (or group of categories) for an individual, 0 otherwise. Such a dummy variable makes it possible to have a pair of potential outcomes and oppose a category (or a group of categories) to another category (or a group of categories). Let E be a non-binary interest variable. We propose a complete disjunctive coding of our variable E. We transform the initial variable to obtain a set of binary vectors (dummy variables), E = (Ee : e ∈ {1, ..., |E|}), where each vector (variable), Ee, takes the value of 0 when its category is not present, and the value of 1 when its category is present, which allows to compute a pairwise-TMLE comparing difference in the outcome between one category and all remaining categories. In order to illustrate the application of our strategy, first, we present the implementation of TMLE to estimate the causal effect of non-binary variable on outcome using simulated data. Secondly, we apply our TMLE adaptation to survey data from the French Political Barometer (CEVIPOF), to estimate the causal effect of education level (A five-level variable) on a potential vote in favor of the French extreme right candidate Jean-Marie Le Pen. Counterfactual reasoning requires us to consider some causal questions (additional causal assumptions). Leading to different coding of E, as a set of binary vectors, E = (Ee : e ∈ {2, ..., |E|}), where each vector (variable), Ee, takes the value of 0 when the first category (reference category) is present, and the value of 1 when its category is present, which allows to apply a pairwise-TMLE comparing difference in the outcome between the first level (fixed) and each remaining level. We confirmed that the increase in the level of education decreases the voting rate for the extreme right party.

Keywords: statistical inference, causal inference, super learning, targeted maximum likelihood estimation

Procedia PDF Downloads 80

Authors: Jaishree Ranganathan, Poonam Rajurkar, Angelina A. Tzacheva, Zbigniew W. Ras

Abstract:

In today’s world, business often depends on the customer feedback and reviews. Sentiment analysis helps identify and extract information about the sentiment or emotion of the of the topic or document. Attribute selection is a challenging problem, especially with large datasets in actionable pattern mining algorithms. Action Rule Mining is one of the methods to discover actionable patterns from data. Action Rules are rules that help describe specific actions to be made in the form of conditions that help achieve the desired outcome. The rules help to change from any undesirable or negative state to a more desirable or positive state. In this paper, we present a Lexicon based weighted scheme approach to identify emotions from customer feedback data in the area of manufacturing business. Also, we use Rough sets and explore the attribute selection method for large scale datasets. Then we apply Actionable pattern mining to extract possible emotion change recommendations. This kind of recommendations help business analyst to improve their customer service which leads to customer satisfaction and increase sales revenue.

Keywords: actionable pattern discovery, attribute selection, business data, data mining, emotion

Procedia PDF Downloads 184