Статьи журнала - International Journal of Information Technology and Computer Science

Все статьи: 1211

Multi-stage Transfer Learning for Fake News Detection Using AWD-LSTM Network

Multi-stage Transfer Learning for Fake News Detection Using AWD-LSTM Network

Sirra Kanthi Kiran, M. Shashi, K. B. Madhuri

Статья научная

In the recent decades, the automatic veracity verification of rumors is essential, since online social media platforms allow users to post news item or express opinion towards a circulating piece of information without much restriction. The intention of fake news is to make the readers believe in inaccurate information, where the detection of fake news by using content is a difficult task. So, the auxiliary information: user profile, social engagement of the users, and other user’s comments are useful in the detection of fake news. In this manuscript, a novel multi-stage transfer learning approach is introduced for an effective fake news detection, where it utilizes user’s comments as auxiliary information to detect whether the given tweet is true or false. The stances of the response tweets contain opinions on news/rumors are often used for verifying the veracity of the circulating information. In order to devastate the effects of the specific rumors at the earliest, the multi-stage transfer learning approach automatically predict veracity of rumors jointly with the stances of their response tweets. The proposed multi-stage transfer learning is an inductive transfer learning variation that is used to forecast the stance of responses, then to identify fake news. The proposed model’s effectiveness is evaluated on the two-benchmark datasets: semEval-2017 task 8 and PHEME. The proposed model outperformed the existing approaches by obtaining a classification accuracy of 64.30% and 65.30%, an F-measure of 65.95% and 63.90% on semEval-2017 task 8, and PHEME on event-wise datasets.

Бесплатно

Multiple Ranks Weighting Score for Microscopic Image Retrieval System

Multiple Ranks Weighting Score for Microscopic Image Retrieval System

P. Suresh, L. Malliga, M. Vijay

Статья научная

Content based medical images have become a major necessity with the growing retrieving Advancements. CBIR access to medical images for supporting clinical decision making has been proposed that would be ease to manage large number of image in the database system. [4] In real time case only few systems has been developed and used in clinical environment. Content-Based Image Retrieval refers to image retrieval system that is based on visual properties of image objects other than textual annotation. Query image features compare with the database image features which is not exactly matching so image feature can be compare with the two tier approach in the database image in order to improve the accuracy of the retrieval system. Every day, large volume of different types of medical images such as MRI, CT images ultrasound, x-ray, radiology, etc are produced in different medical centre’s .microscopic image classification and discrimination (sub-type) [12] is the most difficult problem in medical image retrieval system. In this paper, the survey provides the suitable algorithm for retrieval and classification of medical image to improve the overall accuracy of the MIMS.

Бесплатно

Multiresolution Fuzzy C-Means Clustering Using Markov Random Field for Image Segmentation

Multiresolution Fuzzy C-Means Clustering Using Markov Random Field for Image Segmentation

Xuchao Li, Suxuan Bian

Статья научная

In this paper, an unsupervised multiresolution image segmentation algorithm is put forward, which combines interscale and intrascale Markov random field and fuzzy c-means clustering with spatial constraints. In the initial label determination of wavelet coefficient phase, the statistical distribution property of wavelet coefficients is characterized by Gaussian mixture model, the properties of intrascale clustering and interscale persistence of wavelet coefficients are captured by Markov prior probability model. According to maximum a posterior rule, the initial label of wavelet coefficient from coarse to fine scale is determined. In the image segmentation phase, in order to overcome the shortcomings of conventional fuzzy c-means clustering, such as being sensitive to noise and lacking of spatial constraints, we construct the novel fuzzy c-means objective function based on the property of intrascale clustering and interscale persistence of wavelet coefficients, taking advantage of Lagrange multipliers, the improved objective function with spatial constraints is optimized, the final label of wavelet coefficient is determined by iteratively updating the membership degree and cluster centers. The experimental results on real magnetic resonance image and peppers image with noise show that the proposed algorithm obtains much better segmentation results, such as accurately differentiating different regions and being immune to noise.

Бесплатно

Myanmar-English Bidirectional Machine Translation System with Numerical Particles Identification

Myanmar-English Bidirectional Machine Translation System with Numerical Particles Identification

Yin Yin Win, Aye Thida

Статья научная

This paper the development of Myanmar-English bidirectional machine translation system is implemented applying Rule based machine translation approach. Stanford and ML2KR parsers are used for preprocessing step. From this step, parsers generate corresponding parse tree structures. Used parsers generate corresponding CFG rules which are collected and created as synchronous context free grammar SCFG rules. Myanmar language can be written free order style, but it must be verb final structure. Therefore, CFG rules are required for reordering the structure of the two languages. After that tree to tree transformation is carried on the source tree structure which corresponds with used parser (Stanford parser or ML2KR's parser). When source parse tree is transformed as target parse tree, it is changed according to the SCFG rules. And then system carries out the morphological synthesis. In this stage, we need to solve only for English to Myanmar machine translation because Myanmar language is morphologically rich language. Therefore, particles for Myanmar language can be solved in this system by proposed algorithm. After finishing morphological synthesis, this system generates meaningful and appropriate smoothing sentences.

Бесплатно

Myers-briggs Personality Prediction and Sentiment Analysis of Twitter using Machine Learning Classifiers and BERT

Myers-briggs Personality Prediction and Sentiment Analysis of Twitter using Machine Learning Classifiers and BERT

Prajwal Kaushal, Nithin Bharadwaj B P, Pranav M S, Koushik S, Anjan K Koundinya

Статья научная

Twitter being one of the most sophisticated social networking platforms whose users base is growing exponentially, terabytes of data is being generated every day. Technology Giants invest billions of dollars in drawing insights from these tweets. The huge amount of data is still going underutilized. The main of this paper is to solve two tasks. Firstly, to build a sentiment analysis model using BERT (Bidirectional Encoder Representations from Transformers) which analyses the tweets and predicts the sentiments of the users. Secondly to build a personality prediction model using various machine learning classifiers under the umbrella of Myers-Briggs Personality Type Indicator. MBTI is one of the most widely used psychological instruments in the world. Using this we intend to predict the traits and qualities of people based on their posts and interactions in Twitter. The model succeeds to predict the personality traits and qualities on twitter users. We intend to use the analyzed results in various applications like market research, recruitment, psychological tests, consulting, etc, in future.

Бесплатно

NIPP: Non-Invasive PCOS Prediction using XG-boost Machine Learning Model

NIPP: Non-Invasive PCOS Prediction using XG-boost Machine Learning Model

Shikha Prasher, Leema Nelson, Manal Gafar

Статья обзорная

Polycystic Ovary Syndrome (PCOS) is a common endocrine disorder that affects women of reproductive age, leading to hormonal imbalances and ovarian dysfunction. Early detection and intervention are vital for effective management and prevention of complications. This study compares PCOS prediction using the XGBoost machine learning model against four traditional models: Logistic Regression (LR), Support Vector Machine (SVM), Decision Trees (DT), and Random Forests (RF). LR and SVM achieve accuracies of 95% and 96%, respectively, demonstrating strong predictive capabilities. In contrast, DT had a lower accuracy (82%), indicating limitations in PCOS data complexity. RF showed competitive performance with 96% accuracy, underscoring its effectiveness in ensemble learning. XGBoost achieves 98% accuracy with its parameter configuration. The scale pos weight parameter adjusts the positive class weight in imbalanced datasets, addressing under representation by assigning more weight to the minority class, and thereby improving the training focus. The gradient boosting framework incrementally builds models to address complex feature interactions and dependencies, enhancing the accuracy and stability in predicting intricate PCOS dataset. This analysis highlights the importance of advanced machine learning models such as XGBoost for accurate and reliable PCOS predictions. This research advances PCOS prediction, demonstrates the potential of machine learning in healthcare, and clarifies the strengths and limitations of different algorithms with complex medical datasets.

Бесплатно

New RLS Wiener Smoother for Colored Observation Noise in Linear Discrete-time Stochastic Systems

New RLS Wiener Smoother for Colored Observation Noise in Linear Discrete-time Stochastic Systems

Seiichi Nakamori

Статья научная

In the estimation problems, rather than the white observation noise, there are cases where the observation noise is modeled by the colored noise process. In the observation equation, the observed value y(k) is given as a sum of the signal z(k)=Hx(k) and the colored observation noise v_c(k). In this paper, the observation equation is converted to the new observation equation for the white observation noise. In accordance with the observation equation for the white observation noise, this paper proposes new RLS Wiener estimation algorithms for the fixed-point smoothing and filtering estimates in linear discrete-time wide-sense stationary stochastic systems. The RLS Wiener estimators require the following information: (a) the system matrix for the state vector x(k); (b) the observation matrix H; (c) the variance of the state vector x(k); (d) the system matrix for the colored observation noise v_c(k); (e) the variance of the colored observation noise.

Бесплатно

NoC Research and Practice: Design and Implementation of 2×4 2D-Torus Topology

NoC Research and Practice: Design and Implementation of 2×4 2D-Torus Topology

Xingang Ju, Liang Yang

Статья научная

Design and Implementation of network on chip interconnection architecture for eight compute-intensive processors are mainly presented in this paper. Firstly, it introduces the basic concept and architecture of the NoC, through analysis and comparison of three common NoC topologies, 2×4 2D Turos is chosen as the final topology, and the single routing node architecture is designed, including packet format, routing and arbitration. Secondly, routing nodes coding, routing algorithm and node degree routing direction are designed. Thirdly, the programming and simulation of 2×4 NoC interconnection architecture are designed, and it achieves uninterrupted operation. The result shows the correctness of the interconnection architecture design. Finally, it chooses XC4VSX55-12ff1148 of vertext 4 to synthesize, the maximum frequency can up to 268 MHz, which provides foundation of subsequent research and application.

Бесплатно

Noise and Feedback in Online Communication on Sex: A Study of Nigerian's Conversations on Pornography in Nollywood on Social Networks

Noise and Feedback in Online Communication on Sex: A Study of Nigerian's Conversations on Pornography in Nollywood on Social Networks

F. P. C. Endong

Статья научная

Social networks constitute a suitable forum for debate, and exchange on any sort of topic especially highly sensitive issues. They offer a fertile platform for debate on thorny societal issues such as politics, sex, culture and religion among others. Given the fact that they favor anonymity, openness and non-accountability for voiced opinion, a good number of Nigerians have found them suitable for, "hot", "aggressive" and very passionate discussions over subjects like sex, sexuality and religious convictions – issues which have remarkably remained somehow taboos in the Nigerian society. This paper investigates the conduct of online debates and opinion formation on sex in the prolific Nigerian motion picture industry (Nollywood). It is based on the content analysis of 516 comments by Nigerians, reacting or debating online (in social networks) on pornography in the Nigerian film industry. The paper seeks to explore and quantify the phenomenon of noise in online communication (conservation and debate) on sex by Nigerians. It equally examines how this noise affects communication flow in online debate on pornography in the Nigerian film industry. It argues that being somewhat considerable, noise in such a communication context, is mainly psychological in nature, due principally to the dominance of conservative beliefs on sex and pornography in the Nigerian society. This conservatism motivates most Nigerians to mainly have preconceived stereotypes, notions and biases on sex and pornography and to adopt judgmental and censuring reactions to most attempts to celebrate pornography. The effect of the psychological noise (as observed in online conversation on pornography) is mainly to orchestrate a change of topic from sex to other sensitive issues as politics and religion or engender insults and counter insults which further negatively affect communication.

Бесплатно

Non-polynomial Spline Difference Schemes for Solving Second-order Hyperbolic Equations

Non-polynomial Spline Difference Schemes for Solving Second-order Hyperbolic Equations

Li-Bin Liu, Yong Zhang, Huai-Huo Cao

Статья научная

In this paper, a class of improved methods based on non-polynomial cubic splines in space and finite difference in time direction are constructed for the second-order hyperbolic equations with initial boundary value problems. Truncation error and stability analysis of the methods have been carried out. It is shown that by suitably choosing the parameters, many known methods can be derived from ours. We also obtain a new high accuracy scheme of , which is conditionally stable for .Finally, a numerical experiment is tested and results are compared with other published numerical solutions.

Бесплатно

Novel Approach for Child and Adulthood Classification Based on Significant Prominent Binary Patterns of Local Maximum Edge (SPBPLME)

Novel Approach for Child and Adulthood Classification Based on Significant Prominent Binary Patterns of Local Maximum Edge (SPBPLME)

Rajendra Babu .Ch, Sreenivasa Reddy. E, Prabhakara Rao. B

Статья научная

This paper derives a new procedure for age classification of facial image based on the local region of facial image. The local region of facial image is extracted from a Significant Binary Pattern of Local Maximum Edge (SBPLME). The SBPLME is generated by calculating the absolute value of local difference between the average of local 3×3 sub window pixel values and its neighbors instead of the center pixel value. In the case of Local Maximum Edge Binary Pattern (LMEBP) calculating the absolute value of local difference between the center pixel value of local 3×3 sub window and its neighbors. The proposed SBPLME can generate 512 (0 to 511) different patterns. The present paper utilized Prominent LBP (PLBP) on the proposed SBPLME. The PLBP contains the significant patterns of Uniform LBP (ULBP) and Non Uniform LBP (NULBP). Thus the derived Significant PLBP of Local Maximum Edge (SPBPLME) becomes an efficient image classification and analysis, which will have a significant role in many areas. The novelty of the proposed SPBPLME method is, it has shown excellent age classification results by reducing the overall dimension, thus reducing the overall complexity.

Бесплатно

Novel Hybrid Model: Integrating Scrum and XP

Novel Hybrid Model: Integrating Scrum and XP

Zaigham Mushtaq, M. Rizwan Jameel Qureshi

Статья научная

Scrum does not provide any direction about how to engineer a software product. The project team has to adopt suitable agile process model for the engineering of software. XP process model is mainly focused on engineering practices rather than management practices. The design of XP process makes it suitable for simple and small size projects and not appropriate for medium and large projects. A fine integration of management and engineering practices is desperately required to build quality product to make it valuable for customers. In this research a novel framework hybrid model is proposed to achieve this integration. The proposed hybrid model is actually an express version of Scrum model. It possesses features of engineering practices that are necessary to develop quality software as per customer requirements and company objectives. A case study is conducted to validate the proposal of hybrid model. The results of the case study reveal that proposed model is an improved version of XP and Scrum model.

Бесплатно

Novel Optimized Designs for QCA Serial Adders

Novel Optimized Designs for QCA Serial Adders

A. Mostafaee, A. Rezaei

Статья научная

Quantum-dot Cellular Automata (QCA) is a new and efficient technology to implement logic Gates and digital circuits at the nanoscale range. In comparison with the conventional CMOS technology, QCA has many attractive features such as: low-power, extremely dense and high speed structures. Adders are the most important part of an arithmetic logic unit (ALU). In this paper, four optimized designs of QCA serial adders are presented. One of the proposed designs is optimized in terms of the number of cells, area and delay without any wire crossing methods. Also, two new designs of QCA serial adders and a QCA layout equivalent to the internal circuit of TM4006 IC are presented. QCADesigner software is used to simulate the proposed designs. Finally, the proposed QCA designs are compared with the previous QCA, CNTFET-based and CMOS technologies.

Бесплатно

Novel Spectrum Handoff in Cognitive Radio Networks Using Fuzzy Logic

Novel Spectrum Handoff in Cognitive Radio Networks Using Fuzzy Logic

Nisar A. Lala, Moin Uddin, N.A. Sheikh

Статья научная

Cognitive radio is a technology initiated by many research organizations and academic institutions to raise the spectrum utilization of underutilized channels in order to alleviate spectrum scarcity problem to a larger extent. Spectrum handoff is initiated due to appearance of primary user (PU) on the channels occupied by the secondary user (SU) at that time and location or interference to the PU exceeds the certain threshold. In this paper, we propose a novel spectrum handoff algorithm using fuzzy logic based approach that does two important functions: 1) adjusts transmission power of SU intelligently in order to avoid handoff by reducing harmful interference to PUs and 2) takes handoff decisions intelligently in the light of new parameter such as expected holding time (HT) of the channel as one of its antecedent. Simulated results show impact analysis of selection of the channel in the light of HT information and the comparison with random selection algorithm demonstrates that there is considerable reduction in handoff rate of SU.

Бесплатно

Numerical Implementation of Nonlinear Implicit Iterative Method for Solving Ill-posed Problems

Numerical Implementation of Nonlinear Implicit Iterative Method for Solving Ill-posed Problems

Jianjun Liu, Zhe Wang, Guoqiang He, Chuangang Kang

Статья научная

Many nonlinear regularization methods may converge to local minima in numerical implementation for the complexity of nonlinear operator. Under some not very strict assumptions, we implement our proposed nonlinear implicit iterative method and have a global convergence results. Using the convexity property of the modified Tikhonov functional, it combines nonlinear implicit iterative method with a gradient method for solving ill-posed problems. Finally we present two numerical results for integral equation and parameter identification.

Бесплатно

Off-line sindhi handwritten character identification

Off-line sindhi handwritten character identification

Arsha Kumari, Din Muhammad Sangrasi, Sania Bhatti, Bhawani Shankar Chowdhry, Sapna Kumari

Статья научная

Handwritten Identification is an ability of the computer to receive and translate the intelligible handwritten text into machine-editable text. It is classified into two types based on the way input is given namely: off-line and online. In Off-line handwritten recognition, the input is given in the form of the image while in online input is entered on a touch screen device. The research on off-line and online handwritten Sindhi character identification is on its very initial stage in comparison to other languages. Sindhi is one of the subcontinent's oldest languages with extensive literature and rich culture. Therefore, this paper aims to identify off-line Sindhi handwritten characters. In the proposed work, major steps involve in characters identification are training and testing of the system. Training is performed using a feed-forward neural network based on the efficient accelerative technique, the Back Propagation (BP) learning algorithm with momentum term and adaptive learning rate. The dataset of 304 Sindhi handwritten characters is collected from 16 different Sindhi writers, each with 19 characters. The novelty of proposed work is the comparison of the recognition rate for the single character, two characters and three characters at a time. Results showed that the recognition rate achieved for a single character is more than the recognition rate of multiple characters at a time.

Бесплатно

Offline Handwriting Recognition Using Feedforward Neural Network

Offline Handwriting Recognition Using Feedforward Neural Network

Rosalina, R.B. Wahyu

Статья научная

Many business especially Banks’s services are expanding to include services directed not only to corporate customers but also to individual customer. Furthermore, by the increment of those services, many individual applications to be processed also increases as well. Facing an immense moment, in which requiring more improvements in how it should manage or maintain its applications, some systems or procedures must be improved to match currently increasing customers’ applications. Prior to the improvements, many application forms are filled, input to machine and even to be processed and approved manually. Until recently, application fulfillment processes consists of manual information filling by applicants in an application request paper and later to be re-input by electronic data processing staff which is actually redundant. Aware of such situation, this paper proposes and idea to reduce input processes in an integrated business system by utilizing character recognition system.

Бесплатно

Offline Handwritten Devanagari Script Recognition

Offline Handwritten Devanagari Script Recognition

Ved Prakash Agnihotri

Статья научная

Handwritten Devanagari script recognition system using neural network is presented in this paper. Diagonal based feature extraction is used for extracting features of the handwritten Devanagari script. After that these feature of each character image is converted into chromosome bit string of length 378. More than 1000 sample is used for training and testing purpose in this proposed work. It is attempted to use the power of genetic algorithm to recognize the character. In step-I preprocessing on the character image, then image suitable for feature extraction as here is used. Diagonal based feature extraction method to extract 54 features to each character. In the next step character recognize image in which extracted feature in converted into Chromosome bit string of size 378. In recognition step using fitness function in which find the Chromosome difference between unknown character and Chromosome which are store in data base.

Бесплатно

On Achievable QoS Bounds for Mobile IP through Rayleigh Channel in VANET

On Achievable QoS Bounds for Mobile IP through Rayleigh Channel in VANET

Richa Sharma, Jyoteesh Malhotra

Статья научная

Vehicular Ad Hoc network is a versatile mobile wireless Ad-Hoc network targeted to support traffic monitoring, vehicular safety and many more applications. For the robust and reliable services in the VANET there is need to investigate the performance under frequent handovers in Mobile IP to prevent packet loss. Mobile IP is an interface that helps to track the mobile nodes and deliver messages even if vehicles are out of the coverage area of home node. In order to find the achievable performance bounds in terms of throughput, packet drop, collision rate and packet broadcast rate, extensive simulations have been done. A realistic city scenario has been proposed here by using the Rayleigh channel simulator, mobile IP enabled IEEE 802.11p OBUs and RSUs. The transmission powers of RSUs and threshold power levels have been varied to obtain the optimum performance through realistic conditions. Simulations are performed using NCTUns6.0 (National Chiao Tung University Network Simulator) in mobile IP interface.

Бесплатно

On Approximate Equivalences of Multigranular Rough Sets and Approximate Reasoning

On Approximate Equivalences of Multigranular Rough Sets and Approximate Reasoning

B. K. Tripathy, Anirban Mitra

Статья научная

The notion of rough sets introduced by Pawlak has been a successful model to capture impreciseness in data and has numerous applications. Since then it has been extended in several ways. The basic rough set introduced by Pawlak is a single granulation model from the granular computing point of view. Recently, this has been extended to two types of multigranular rough set models. Pawlak and Novotny introduced the notions of rough set equalities which is called approximate equalities. These notions of equalities use the user knowledge to decide the equality of sets and hence generate approximate reasoning. However, it was shown by Tripathy et al, even these notions have limited applicability to incorporate user knowledge. So the notion of rough equivalence was introduced by them. The notion of rough equalities in the multigranulation context was introduced and studied. In this article, we introduce the concepts of multigranular rough equivalences and establish their properties. Also, the replacement properties, which are obtained by interchanging the bottom equivalences with the top equivalences, have been established. We provide a real life example for both types of multigranulation, compare the rough multigranular equalities with the rough multigranular equivalences and illustrate the interpretation of the rough equivalences through the example.

Бесплатно

Журнал