Browsing by Author "Arman, Md. Shohel"

Now showing 1 - 18 of 18

A Comparative Analysis of Four Classification Algorithms for University Students Performance Detection
(Lecture Notes in Electrical Engineering, Springer, 2020-03-24) Das, Dipta; Shakir, Asif Khan; Rabbani, Md. Shah Golam; Rahman, Mostafijur; Shaharum, Syamimi Mardiah; Khatun, Sabira; Fadilah, Norasyikin Binti; Qaiduzzaman, Khandker M.; Islam, Md. Shariful; Arman, Md. Shohel
The student’s performance plays an important role in producing the best quality graduate who will responsible for the country’s economic growth and social development. The labor market also concerns with student’s performance because the fresh graduate students are considered as an employee depends on their academic performance. So, identification of the reason behind student’s performance variation provides valuable information for planning education and policies. Many researchers try to find out the reason with different types of data mining approaches in different countries. However, none of them worked with Bangladeshi students. This paper proposed a model for identifying the key factors of variation Bangladeshi students’ academic performance and predicts their results. This paper proposes a model which able to identify the students who need special attention. Different types of feature selection methods were used such as Co-relation, Chi-Square and Euclidean distance to select valuable features and feature selections result through decision tree, Naive Bayes, K-nearest neighbor and Artificial Neural Network classifiers algorithm were compared. The performance analysis is done by using student SGPA and review on given facilities from a university. From the performance analysis result it is found that, decreasing number of classes in dataset, the Artificial Neural Network (ANN) (93.70%) performs better than Decision Tree (DT) (92.18%), K-Nearest Neighbors (KNN) (77.74%) and Naïve Bayes (NB) (68.33%). However, an increasing number of classes in dataset the DT perform better than ANN, KNN, NB.
A Machine Learning-Based Approach for Sentiment Analysis of Movie Reviews on a Bangladeshi OTT Platform
(Springer, 2023-12-15) Jahan, Hasnur; Arman, Md. Shohel; Hasan, Afia; Bristy, Sabikun Nahar
Sentiment analysis is the examination of feelings and viewpoints in any kind of literature. Opinion mining is another phrase for sentiment analysis. The data’s sentiment analysis is quite helpful, to convey the collective, group, or individual viewpoint. This method is employed to ascertain a person’s attitude about a specific source. Huge amounts of data are present on social media and other online platforms in the form of tweets, blogs, status, postings, etc. The movie reviews were examined in this research using a variety of methods. On demand of movies on OTT platform several Facebook reviewer pages has been created in Bangladesh. For this work, almost 1000 Bangla reviews were gathered, containing some English word from Facebook. Customer tones were assessed in movie reviews. We use Unigram, Bigram, and Trigram features with a variety of models, including Decision Tree, Random Forest, Multinomial Naive Bayes, K-Neighbors, and Linear Support Vector Machine in n-grams. Random Forest is the most accurate, with 92.35 percent and 90.03 percent accuracy in Unigram and Trigram, respectively. The most accurate model in Bigram is Decision Tree, which has an accuracy of 89.50 percent. This proposed system will help to analysis reviews and give feedback about a movie.
A Web Based Four-Tier Architecture using Reduced Feature Based Neural Network Approach for Prediction of Student Performance
(Scopus, 2021) Hossen, Md. Anwar; Alamgir, Rakib Bin; Alam, Arman Ul; Siddika, Fatema; Hossain, Shah Fahad; Arman, Md. Shohel
Enhancing student's performance is a significant part of developing quality education in any educational institute. It is very difficult to get promising student performance without student categorization according to their academic performance as there are different standardized students. In this paper, our aim is to determine the performance of the students. For this purpose, a survey has been conducted on students in our university in order to collect data and to analyze and predict the student category based on their performance. Apart from this, another purpose of this study is to examine the effect of the reduced features on the classification model using state-of-art machine learning algorithms. Here, we propose a workflow of web-based four-tier architecture for the student performance prediction that will define the student's category in order to help them exactly pinpoint their learning capabilities. Hence, we used multiple supervised learning-based machine learning techniques for the prediction of student performance. Each of the student category categorized by considering on the top features. The analysis results indicate that we got the highest performance that is 88.00% by using the Artificial Neural Network (ANN) among the classifiers by showing its superiority to the existing model.
A Web Based Four-Tier Architecture using Reduced Feature Based Neural Network Approach for Prediction of Student Performance
(IEEE, 2021-01) Hossen, Md. Anwar; Bin Alamgir, Rakib; Alam, Arman Ul; Siddika, Fatema; Hossain, Shah Fahad; Arman, Md. Shohel
Enhancing student's performance is a significant part of developing quality education in any educational institute. It is very difficult to get promising student performance without student categorization according to their academic performance as there are different standardized students. In this paper, our aim is to determine the performance of the students. For this purpose, a survey has been conducted on students in our university in order to collect data and to analyze and predict the student category based on their performance. Apart from this, another purpose of this study is to examine the effect of the reduced features on the classification model using state-of-art machine learning algorithms. Here, we propose a workflow of web-based four-tier architecture for the student performance prediction that will define the student's category in order to help them exactly pinpoint their learning capabilities. Hence, we used multiple supervised learning-based machine learning techniques for the prediction of student performance. Each of the student category categorized by considering on the top features. The analysis results indicate that we got the highest performance that is 88.00% by using the Artificial Neural Network (ANN) among the classifiers by showing its superiority to the existing model.
Aspect Based Sentiment Analysis in Bangla Dataset Based on Aspect Term Extraction
(Springer, 2020-07-30) Haque, Sabrina; Rahman, Tasnim; Shakir, Asif Khan; Arman, Md. Shohel; Biplob, Khalid Been Badruzzaman; Himu, Farhan Anan; Das, Dipta; Islam, Md Shariful
Recent years have seen rapid growth of research on sentiment analysis. In aspect-based sentiment analysis, the idea is to take sentiment analysis a step further and find out what exactly someone is talking about, and then measuring the sentiment if she or he likes or dislikes it. Sentiment analysis in Bengali language is progressing and is considered as an important research interest. Due to scarcity of resources like proper annotated dataset, corpora, lexicon such as part of speech tagger etc. aspect-based sentiment analysis hardly has been done in Bengali language. In this paper, we have conducted our experiments based on a recent work from 2018 using conventional supervised machine learning algorithms (RF, SVM, KNN) to perform one of the ABSA’s tasks - aspect category extraction. The work is done on two datasets named – Cricket and Restaurant. We then compared our results with the existing work. We used two traditional steps to clean data and found that less preprocessing leads to better F1 Score. For Cricket dataset, SVM and KNN performed better, resulting F1 score of 37% and 27%. For Restaurant dataset, RF and SVM achieved improved score of 35% and 39% respectively. Additionally, we selected two more algorithms LR and NB, LR achieved best F1 score (43%) for Restaurant dataset among all.
Assessing the Effectiveness of Topic Modeling Algorithms in Discovering Generic Label with Description
(Springer, 2020-02-13) Rahman, Shadikur; Hossain, Syeda Sumbul; Arman, Md. Shohel; Rawshan, Lamisha; Toma, Tapushe Rabaya; Rafiq, Fatama Binta; Md. Badruzzaman, Khalid Been
Analyzing short text or documents using topic modeling becomes a popular solutions for the increasing number of documents produced in everyday life. For handling the large amount of documents, many topic modeling algorithms are used e.g. LDA, LSI, pLSI, NMF. In this study, we have used LDA, LSI, NMF and also lexical database wordNet synset for candidate labels in our topics labeling. And finally compare the effectiveness of topic modeling algorithms for short documents. Among those LDA gives the better result in terms of WUP similarity. This study will help to select the proper algorithm for labeling topics and can easily identify the meaning of topics.
Bangladeshi Stock Price Prediction and Analysis with Potent Machine Learning Approaches
(Springer, 2020-07-30) Das, Sajib; Arman, Md. Shohel; Hossain, Syeda Sumbul; Islam, Md. Sanzidul; Himu, Farhan Anan; Shakir, Asif Khan
Stock price forecasting, is one of the most significant financial complexities, since data are not reliable and noisy, impacting many factors. This article offers a machine learning model for the stock price prediction using Support Vector Machine-Regression (SVR) with two different kernels which are Radial Basis Function (RBF) and linear kernel. This study shows the Prediction and accuracy comparison between Support Vector Regression (SVR) and Linear Regression (LR) and also the accuracy comparison for different kernels of Support vector Regression (SVR). The model has used sum squared error (SSE) to determine the accuracy of each algorithm; which has shown significant improvement than the other studies. This analysis is conducted on the price data of about five years of Grameenphone listed on Dhaka Stock Exchange (DSE). The highest accuracy was found with Linear Regression model in every case with the highest accuracy of about 97.07% followed by SVR (Linear) model and SVR (radial basis function) model with the highest accuracy rate of about 97.06% and 96.82%. In some cases the accuracy of SVR (radial basis function) was higher than SVR (linear). But it was the Linear Regression which had the highest accuracy of all in every case.
Classification of Immunity Booster Medicinal Plants Using CNN
(Scopus, 2021) Musa, Md.; Arman, Md. Shohel; Hossain, Md. Ekram; Thusar, Ashraful Hossen; Nisat, Nahid Kawsar; Islam, Arni
Environment has blessed us with various kinds of plants. Some of them uses as resources of medicines as it is called medicinal plant. In Bangladesh medicinal plants are also known as Ayurveda, Homeopathy and Unani. Experts says medicinal plants can be very useful in the fight with recent pandemic which is Covid-19. As we know health of a body depends on its immune system, so it is important to keep immunity stronger. Strong immune system can be influential to any infectious virus, bacteria and pathogens. On the other hand, inactive one can get easily infected with virus and other illness. There are certain medicinal plants which reinforce our immunity. Therefore, classification of these medical plants is very important. For this classification we have collected leaf images for six different classes which’s local names are Darchini, Tulshi, Tejpata, Sojne, Neem, Pathorkuchi. In this article we introduced a famous algorithm for classification named CNN (Convolutional neural network). We used CNN (Convolutional neural network) to recognize the plant from leaf images and got 95.58% accuracy. In future infectious virus can appear which can be more threatening than others, our research will help people to know about immune system and medicinal plants which reinforce our immunity, so that they can fight with diseases and viruses.
Classification of Immunity Booster Medicinal Plants Using CNN
(Scopus, 2021) Musa, Md.; Arman, Md. Shohel; Hossain, Md. Ekram; Thusar, Ashraful Hossen; Nisat, Nahid Kawsar
Environment has blessed us with various kinds of plants. Some of them uses as resources of medicines as it is called medicinal plant. In Bangladesh medicinal plants are also known as Ayurveda, Homeopathy and Unani. Experts says medicinal plants can be very useful in the fight with recent pandemic which is Covid-19. As we know health of a body depends on its immune system, so it is important to keep immunity stronger. Strong immune system can be influential to any infectious virus, bacteria and pathogens. On the other hand, inactive one can get easily infected with virus and other illness. There are certain medicinal plants which reinforce our immunity. Therefore, classification of these medical plants is very important. For this classification we have collected leaf images for six different classes which’s local names are Darchini, Tulshi, Tejpata, Sojne, Neem, Pathorkuchi. In this article we introduced a famous algorithm for classification named CNN (Convolutional neural network). We used CNN (Convolutional neural network) to recognize the plant from leaf images and got 95.58% accuracy. In future infectious virus can appear which can be more threatening than others, our research will help people to know about immune system and medicinal plants which reinforce our immunity, so that they can fight with diseases and viruses.
CryptoAR: Scrutinizing the Trend and Market of Cryptocurrency Using Machine Learning Approach on Time Series Data
(Scopus, 22-11) Bitto, Abu Kowshir; Mahmud, Imran; Bijoy, Md. Hasan Imam; Jannat, Fatema Tuj; Arman, Md. Shohel; Shohug, Md. Mahfuj Hasan; Jahan, Hasnur
Cryptocurrencies are encrypted digital or virtual money used to avoid counterfeiting and double spending. The scope of this study is to evaluate cryptocurrencies and forecast their price in the context of the currency rate trends. A public survey was conducted to determine which cryptocurrency is the most well-known among Bangladeshi people. According to the survey respondents, Bitcoin is the most famous cryptocurrency among the eight digital currencies. After that, we'll explore the four most well-known cryptocurrencies: Bitcoin, Ethereum, Litecoin, and Tether token. The 'YFinance' python package collects our cryptocurrency dataset, and the relative strength index (RSI) is employed to investigate these cryptocurrencies. Autoregressive (AR), moving average (MA), and autoregressive moving average (ARMA) models are applied to our time-series data from 2015-1-1 to 2021-6-1. Using the 'closing' price and a simple moving average (SMA) graph, bitcoin and tether are identified as oversold or overbought cryptocurrencies. We employ the seasonal decomposed technique into the dataset before implementing the model, and the augmented dickey-fuller test (ADF) indicates too much seasonality in the dataset. The autoregressive (AR) model is the most accurate in predicting the price of Bitcoin, Ethereum, Litecoin, and Tether-token, with 97.21%, 96.04%, 95.8%, and 99.91% accuracy, consecutively
Detection and Classification of Road Damage Using R-CNN and Faster R-CNN
(Lecture Notes of the Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering, Springer, 2020-07-30) Arman, Md. Shohel; Hasan, Md. Mahbub; Sadia, Farzana; Shakir, Asif Khan; Sarker, Kaushik; Himu, Farhan Anan
Road surface monitoring is mostly done manually in cities which is an intensive process of time consuming and labor work. The intention of this paper is to research on road damage detection and classification from road surface images using object detection method. This paper applied multiple convolutional neural network (CNN) algorithm to classify road damage and discovered which algorithm performs better in road damage detection and classification. The damages are classified in three categories pothole, crack and revealing. For this research data was collected from street of Dhaka city using smartphone camera and prepossessed the data like image resize, white balance, contrast transformation, labeling. This study applies R-CNN and faster R-CNN for object detection of road damages and apply Support Vector Machine (SVM) for classification and gets a better result from previous studies. Then losses are calculated using different loss functions. The results demonstrate the highest 98.88% accuracy and the lowest loss is 0.01.
Distributed Database Problems, Approaches and Solutions
(International Journal of Machine Learning and Computing, 2018-10-05) Rana, Md. Shohel; Sohel, Mohammad Khaled; Arman, Md. Shohel
The distributed database system is the combination of two fully divergent approaches to data processing: database systems and computer network to deliver transparency of distributed and replicated data. The key determination of this paper is to achieve data integration and data distribution transparency, study and recognize the problems and approaches of the distributed database system. The distributed database is evolving technology to store and retrieve data from several location or sites with maintaining the dependability and obtainability of the data. In the paper we learn numerous problems in distributed database concurrency switch, design, transaction management problem etc. Distributed database allows to end worker to store and retrieve data anywhere in the network where database is located, during storing and accessing any data from distributed database through computer network faces numerous difficulties happens e.g. deadlock, concurrency and data allocation using fragmentation, clustering with multiple or single nodes, replication to overcome these difficulties it is essential to design the distributed database sensibly way.
Measuring the Effectiveness of Software Code Review Comments
(Communications in Computer and Information Science, Springer, 2020-07-18) Hossain, Syeda Sumbul; Arafat, Yeasir; Hossain, Md. Ekram; Arman, Md. Shohel; Islam, Anik
Code reviewing becomes a more popular technique to find out early defects in source code. Nowadays practitioners are going for peer reviewing their codes by their co-developers to make the source code clean. Working on a distributed or dispersed team, code review is mandatory to check the patches to merge. Code reviewing can also be a form of validating functional and non-functional requirements. Sometimes reviewers do not put structured comments, which becomes a bottle neck to developers for solving the findings or suggestions commented by the reviewers. For making the code review participation more effective, structured and efficient review comments is mandatory. Mining the repositories of five commercialized projects, we have extracted 15223 review comments and labelled them. We have used 8 different machine learning and deep learning classifiers to train our model. Among those Stochastic Gradient Descent (SGD) technique achieves higher accuracy of 80.32%. This study will help the practitioners to build up structured and effective code review culture among global software developers.
Modelling the Knowledge-Sharing Behaviour of Students on Facebook
(ProQuest LLC, 2018) Hossain, Mohamed Emran; Bhuiyan, Touhid; Mahmud, Imran; Arman, Md. Shohel
This paper aims to illustrate the relationship between the constructs of social cognitive theory and social exchange theory with regard to the knowledge-sharing behaviour of students on Facebook. This research was conducted on 123 students using self-administrative survey questionnaires. The technique of structural equation modelling was employed to examine the hypothesized relationships between the variables. The findings of this study indicate that affiliation and innovativeness significantly the knowledge-sharing behaviour of students. Overall, perceived reciprocal benefit, perceived enjoyment, knowledge power, and affiliation and outcome expectations are found to be strong predictors of such behaviour. Previous research mostly examined the knowledge sharing attitude or intention in the industry setting. This study has been conducted in the educational setting and particularly focuses on the influence of the educational climate and expectation outcome on the knowledge sharing attitude of students.
Modelling Turn Away Intention of Information Technology Professionals in Bangladesh
(International Journal of Electrical and Computer Engineering, 2020-10-05) Arman, Md. Shohel; Akter, Rozina; Mahmud, Imran; Ramayah, T.
Despite, Bangladesh produces many IT graduates each year but only one tenth of total graduates contribute in IT development sector. In order to keep the contribution to economy through IT development, it is crucial for IT industry to know the factors that influence turn away of IT graduates. In this paper, building upon role stress theory, we develop a research model to explore the influence of workplace exhaustion and threat of professional obsolescence (TPO). Data were gathered from 185 IT professionals from 15 different IT companies through survey questionnaire. The structural equation modelling technique was used to test the paths. The results suggests that strong influence of TPO on turn-away intentions. Result also suggests significant roles of work overload, family-career conflict and control over career and workplace exhaustion on turn away intention. This paper contributes to the body of work dedicated to helping us better understand the turn away behaviour from the workplace exhaustion and TPO perspectives. From the viewpoint of practice, this research sheds light on some of the challenges that the IT industry might face when making strategy and policy to control turn away from IT profession in Bangladesh
My knowledge is not enough: An investigation on the impact of threat of professional obsolescence on turn away intention among IT professionals in Bangladesh
(Researchgate, 2017) Arman, Md. Shohel; Mahmud, Imran; Ramayah, T.; Rabaya, Tapushe; Rawshon, Shahriar
While the economy of Bangladesh is booming in information technology (IT) sector, reports suggest that only 10% IT or computer science graduates are contributing in this development. IT companies are challenged to hire new IT graduates as many professionals are switching job to teaching or management positions or moving aboard for higher studies. Using the work place exhaustion and threat of professional obsolescence (TPO) as theoretical lenses, this study explains what are the factors negatively impacts turn away intention of IT professionals. Specifically, we hypothesize that threat of professional obsolescence is very crucial and impacts very high on turn away intention. Data were collected from 185 IT professionals from 15 different IT companies in Bangladesh by survey questionnaire. Structural equation modeling technique was used to analyze the data and test the research model. The results of testing the model indicate the central role of TPO on turn away intention. This study advances the theoretical understanding on turn away intention and offers suggestions for both academia and organizations. The paper also acknowledges the limitations of the study and suggests research directions for future researchers.
Predicting the Death of Road Accidents in Bangladesh Using Machine Learning Algorithms
(Scopus, 2021) Siddik, Md. Abu Bakkar; Arman, Md. Shohel; Hasan, Afia; Jahan, Mahmuda Rawnak; Islam, Majharul; Biplob, Khalid Been Badruzzaman
Road accidents are now a common occurrence in our country. Every year thousands of people die in these accidents and thousands of people are crippled and cursed. Recently the level of road accidents has increased drastically. In this research paper, the authors discuss previous road accident history profoundly and predict death by applying the machine learning algorithm to get appropriate accuracy in Bangladesh. In this study, we had applied four classification models such as Decision Tree, K-Nearest Neighbors (KNN), Naïve Bayes and Logistic Regression to predict the death of road accidents in Bangladesh. The model was constructed, trained, and tested using the data from “Prothom Alo” newspaper, from which we collected 1237 road crash incidents. This research would be helpful for the policymakers and stakeholders related to the road to take the future steps with the highest accuracy of 88% in the Decision tree algorithm.
Sentiment Analysis From Bangladeshi Food Delivery Startup Based on User Reviews Using Machine Learning and Deep Learning
(Institute of Advanced Engineering and Science (IAES), 2023-08-15) Bitto, Abu Kowshir; Bijoy, Md. Hasan Imam; Arman, Md. Shohel; Mahmud, Imran; Das, Aka; Majumder, Joy
Food delivery methods are at the top of the list in today's world. People's attitudes toward food delivery systems are usually influenced by food quality and delivery time. We did a sentiment analysis of consumer comments on the Facebook pages of Food Panda, HungryNaki, Pathao Food, and Shohoz Food, and data was acquired from these four sites’ remarks. In natural language processing (NLP) task, before the model was implemented, we went through a rigorous data pre-processing process that included stages like adding contractions, removing stop words, tokenizing, and more. Four supervised classification techniques are used: extreme gradient boosting (XGB), random forest classifier (RFC), decision tree classifier (DTC), and multi nominal Naive Bayes (MNB). Three deep learning (DL) models are used: convolutional neural network (CNN), long term short memory (LSTM), and recurrent neural network (RNN). The XGB model exceeds all four machine learning (ML) algorithms with an accuracy of 89.64%. LSTM has the highest accuracy rate of the three DL algorithms, with an accuracy of 91.07%. Among ML and DL models, LSTM DL takes the lead to predict the sentiment.