Browsing by Author "Mahmud, Imran"

Now showing 1 - 20 of 79

A Deep Learning Study on Understanding Banglish and Abbreviated Words Used in Social Media
(5th International Conference on Intelligent Computing and Control Systems (ICICCS), IEEE, 2021-05-26) Das, Suharto; Islam, Md Sanzidul; Mahmud, Imran
Day by Day, the trend of using social media(SM) has increased among the people. With the increasing rate, the use of Banglish (Merge of Bangla and English) and shortcut words has also being increased. In this research, Banglish and Shortcut words have been fully converted to English. For this type of conversation, we have used the CNNs method and the method consist of multiple layer and these layers are connected to each other. This method is considered to be the best method because it does not need any feature extraction. The convolutional neuron network is used in many areas such as image and pattern recognition, speech recognition, natural language processing and video analysis. In this research we have used CNN method because we have decided to use computer vision as some of the words are close to each other and at first we need to convert all the shortcut words into images for pre-processing the data. With the help of CNN method searched for the Banglish and shortcut words that people use for their daily conversation. At first we find the different representative of a single word and then we converted those shortcut and Banglish words to main word using CNNs method.
A Machine Learning Approach for Risk Factors Analysis and Survival Prediction of Heart Failure Patients
(Elsevier, 2023-04-18) Ali, Md. Mamun; Al-Doori, Vian S.; Mirzah, Nubogh; Hemu, Asifa Afsari; Mahmud, Imran; Azam, Sami; Al-tabatabaie, Kusay Faisal; Ahmed, Kawsar; Bu, Francis M.; Moni, Mohammad Ali
In this study, we propose machine learning (ML) for risk factors analysis and survival prediction of Heart Failure (HF) patients using a survival dataset. Five supervised ML methods are applied to the dataset: Decision Tree (DT), Decision Tree Regressor (DTR), Random Forest (RF), XGBoost, and Gradient Boosting (GB) algorithms. We compare the applied algorithms’ performances based on accuracy, precision, recall, F-measure, and log loss value and show RF provides the highest accuracy of 97.78%. The analysis of the risk factors shows the most predictive features based on coefficients and feature importance. The top six risk factors for HF patients are serum creatinine (SC), age, ejection fraction (EF), platelets, creatinine phosphokinase (CPK), and SS (SS). Further analysis of these factors shows significant clustering of the features. The survival analysis finds that the increment of SC, age, and SS and the decrement of EF are the most significant risk factors for HF patients. Our results suggest that HF survival prediction is possible with higher accuracy using the proposed model. Our ML models are useful in clinical settings for screening patients with HF probability.
A Novel Front Door Security (FDS) Algorithm Using GoogleNet-BiLSTM Hybridization
(IEEE, 2023-02-23) Paula, Luiz Paulo Oliveira; Faruqui, Nuruzzaman; Mahmud, Imran; Whaiduzzaman, Md.; Hawkinson, Eric Charles; Trivedi, Sandeep
Security has always been a significant concern since the dawn of human civilization. That is why we build houses to keep ourselves and our belongings safe. And we do not hesitate to spend a lot on front-door locks and install CCTV cameras to monitor security threats. This paper presents an innovative automatic Front Door Security (FDS) algorithm that uses Human Activity Recognition (HAR) to detect four different security threats at the front door from a real-time video feed with 73.18% accuracy. The activities are recognized using an innovative combination of GoogleNet-BiLSTM hybrid network. This network receives the video feed from the CCTV camera and classifies the activities. The proposed algorithm uses this classification to alert any attempts to break the door by kicking, punching, or hitting. Furthermore, the proposed FDS algorithm is effective in detecting gun violence at the front door, which further strengthens security. This Human Activity Recognition (HAR)-based novel FDS algorithm demonstrates the potential of ensuring better safety with 71.49% precision, 68.2% recall, and an F1-score of 0.65.
A Predictive Analysis Framework of Heart Disease Using Machine Learning Approaches
(Daffodil International University, 22-06-29) Molla, Shourav; Shamrat, F. M. Javed Mehedi; Rafi, Raisul Islam; Umaima, Umme; Umaima, Umme; Hossain, Shahed; Mahmud, Imran
Heart disease is among the leading causes for death globally. Thus, early identification and treatment are indispensable to prevent the disease. In this work, we propose a framework based on machine learning algorithms to tackle such problems through the identification of risk variables associated to this disease. To ensure the success of our proposed model, influential data pre-processing and data transformation strategies are used to generate accurate data for the training model that utilizes the five most popular datasets (Hungarian, Stat log, Switzerland, Long Beach VA, and Cleveland) from UCI. The univariate feature selection technique is applied to identify essential features and during the training phase, classifiers, namely extreme gradient boosting (XGBoost), support vector machine (SVM), random forest (RF), gradient boosting (GB), and decision tree (DT), are deployed. Subsequently, various performance evaluations are measured to demonstrate accurate predictions using the introduced algorithms. The inclusion of Univariate results indicated that the DT classifier achieves a comparatively higher accuracy of around 97.75% than others. Thus, a machine learning approach is recognize, that can predict heart disease with high accuracy. Furthermore, the 10 attributes chosen are used to analyze the model's outcomes explain ability, indicating which attributes are more significant in the model's outcome.
An Analysis on Breast Disease Prediction Using Machine Learning Approaches
(International Journal of Scientific and Technology Research, 2020) Shamrat, F. M. Javed Mehedi; Raihan, Md. Abu; Rahman, A.K.M. Sazzadur; Mahmud, Imran; Akter, Rozina
The central aspect of this study is to evaluate the different Machine learning classifier's performance for the prediction of breast cancer disease. In this work, we have used six supervised classification techniques for the classification of breast cancer disease. For example, SVM, NB, KNN, RF, DT, and LR used for the early prediction of breast cancer. Therefore, we evaluated breast cancer dataset through sensitivity, specificity, f1 measure, and total accuracy. The prediction performance of breast cancer analysis shows that SVM obtained the uppermost performance with the utmost classification accuracy of 97.07%. Whereas, NB and RF have achieved the second highest accuracy by prediction. Our findings can help to reduce the existence of breast cancer disease through developing a machine learning-based predictive system for early prediction.
An Investigation on Exhaustion of SAP ERP Users: Influence of Pace of Change and Technostress
(Semantic Scholar, 2017-10-01) Roy, Prashanta Kumar; Mahmud, Imran; Jahan, Nusrat; Sadia, Farzana
Despite recent growing research interest on ERP research, the understanding on ERP induced exhaustion is still limited. This study examines how the pace of change of ERP functionalities and interface causes exhaustion in workplace. For this purpose, we conducted an investigation on 128 ERP users from two different organizations in Bangladesh. We extended theory of technostress by integrating pace of change of ERP system. Result suggests that pace of change on ERP system significantly affect work-overload, work-life conflict and role ambiguity on ERP users. Result also shows that work-overload and role ambiguity are strong predictors for ERP induced exhaustion.
An SMS and web-based traffic case management system in Bangladesh
(BRAC University, 2006-08) Mahmud, Imran; Roy, Premankur; Saha, Sangeet; Salam, Sayeed
The Traffic Police of Bangladesh has no database software where the cases can be stored. For this reason the Traffic Police update all the cases manually. They have a case record book. If any incident occurs and the police enter any cases against that vehicle or driver then it will be stored in that case book. This process costs much time and it is not an efficient process. We had studied about this and proposed an efficient Traffic case management system in Bangladesh. In our proposed system there will have an official web-based software, offline software and central database. In every police box there will have a computer where webbased software will be installed. If the traffic police want to enter a case against anybody then he can update it through the web-based software. It will be stored in the central database. Any police can see the updated information from any police box within a minute. So it is very time consuming. There will have also offline software, which will also be connected with the central database. If for any reason wed-based software is damaged then offline software will work. In our proposed system there is an SMS system also. If any police wants to know the information about any car, driver or case then he/she will SMS the vehicle number to a particular number. After some moment the police will receive an SMS where all the information about the vehicle will be shown. This information will come from central database. So it will be more efficient process than any other existing system in Bangladesh.
Analysing Most Efficient Deep Learning Model To Detect COVID-19 From Computer Tomography Images
(Daffodil International University, 2022-09-16) Shamrat, F. M. Javed Mehedi; Chakraborty, Sovon; Ahammad, Rasel; Shitab, Tanzil Mahbub; Kazi, Md. Aslam; Hossain, Alamin; Mahmud, Imran
COVID-19 illness has a detrimental impact on the respiratory system, and the severity of the infection may be determined utilizing a selected imaging technique. Chest computer tomography (CT) imaging is a reliable diagnostic technique for finding COVID-19 early and slowing its progression. Recent research shows that deep learning algorithms, particularly convolutional neural network (CNN), may accurately diagnose COVID-19 using lung CT scan images. But in an emergency, detection accuracy simply is not enough. Determinants of data loss and classification completion time play a critical element. This study addresses the issue by finding the most efficient CNN model with the least data loss and classification time. Eight deep learning models, including Max Pooling 2D, Average Pooling 2D, VGG19, VGG16, MobileNetV2, InceptionV3, AlexNet, NFNet using a dataset of 16000 CT scans image data of COVID-19 and non-COVID-19 are compared in the study. Using the confusion matrix, the performance of the models is compared and together with the data loss and completion time. It is observed from the research that MobileNetV2 provides the highest accurate result of 99.12% with the least data loss of 0.0504% in the lowest classification completion time of 16.5secs per epoch. Thus, employing MobileNetV2 gives the best and the quickest result in an emergency.
Application of K-Means Clustering Algorithm to Determine the Density of Demand of Different Kinds of Jobs
(International Journal of Scientific and Technology Research, 2020) Shamrat, F. M. Javed Mehedi; Tasnim, Zarrin; Mahmud, Imran; Jahan, Ms. Nusrat; Nobel, Naimul Islam
In the current competitive job market, information is the most powerful tool. As a job, the seeker looks for a job, and he must have the insight of what kind of competition he is about to face. This information will allow the job seeker to improve himself from the rest in the market. To determine the demand for any field of job among job seekers, with the help of the unsupervised k-means machine learning algorithm, the data of job interests can be clustered in different groups based on their kinds. The visual representation of the clusters in a scatter plot gives the information on which variety of jobs are in more or less demand among job seekers with the density of the groups. This study provides insight into the current jobmarket.
Approach of Different Classification Algorithms to Compare in N-gram Feature between Bangla Good and Bad Text Discourses
(Springer, 2023-05-31) Bitto, Abu Kowshir; Bijoy, Md. Hasan Imam; Khan, Saima; Mahmud, Imran; Biplob, Khalid Been Badruzzaman
Bangla Natural Language Processing (BNLP) is a newish challenge in Artificial Intelligence. With the rapid expansion of the Bangla language, it is now adopted on a variety of platforms, including social media, communication platforms, news media, and so on. The classification of text documents becomes an important factor in resolving the challenge of information organization and knowledge management. This study uses five supervised classification methods to explore the categorization of Bangla text discourse using N-gram (unigram, bigram, and trigram) features. Bangla text discourse is collected from different platforms such as social media, personal Bangla blogs, and people's utterances in order to accomplish the research goal. After collecting data, the most difficult part of the Bangla language preprocessing is completed, which includes adding contractions, removing punctuations, encoding, and a variety of other operations. For this study, 1499 text documents were initially used, with 1459 Bangla text discourses being used after preprocessing. To convert the text into a token, N-gram feature methods utilizing TF-IDF-Vectorizer are used. During the experiment phase, unigram, bigram, and trigram feature techniques are used to apply Logistic Regression (LR), Decision Tree Classifier (DTC), Random Forest (RF), Multinomial Naive Bayes (MNB), and K-Nearest Neighbors (KNN) models to the dataset. In the unigram and bigram features, Multinomial Naive Bayes (MNB) outperformed all other classifiers, with the highest accuracy of 89.31% and 86.94%, respectively. The trigram feature of K-Nearest Neighbors (KNN) achieves a maximum accuracy of 84.25%, and the proposed model can classify the Bangla text document as Good or Bad Discourse.
Bitcoin Trading Indicator: A Machine Learning Driven Real Time Bitcoin Trading Indicator for the Crypto Market
(Institute of Advanced Engineering and Science (IAES), 2023-06-18) Rahaman, Ashikur; Bitto, Abu Kowshir; Biplob, Khalid Been Md. Badruzzaman; Bijoy, Md. Hasan Imam; Jahan, Nusrat; Mahmud, Imran
As opposed to other fiat currencies, bitcoin has no relationship with banks. Its price fluctuation is largely influenced by fresh blocks, news, mining information, support or resistance levels, and public opinion. Therefore, a machine-learning model will be fantastic if it learns from data and tells or indicates if we need to purchase or sell for a little period. In this study, we attempted to create a tool or indicator that can gather tweets in real-time using tweepy and the Twitter application programming interface (API) and report the sentiment at the time. Using the renowned Python module "FBProphet," we developed a model in the second phase that can gather historical price data for the bitcoin to US dollar (BTCUSD) pair and project the price of bitcoin. In order to provide guidance for an intelligent forex trader, we finally merged all of the models into one form. We traded with various models for a very little number of days to validate our bitcoin trading indicator (BTI), and we discovered that the combined version of this tool is more profitable. With the combined version of the instrument, we quickly and with little error root mean square error (RMSE: 1,480.58) generated a profit of $1,000.71 USD.
Civic engagement through restaurant review page in Facebook
(International Journal of Ethics and Systems, Emerald, 2021-01-18) Satter, A.K.M. Zaidi; Mahmud, Arif; Rahman, Ashikur; Mahmud, Imran; Akter, Rozina
Purpose Existing literature affirms that almost half of the young generation has remained unemployed worldwide. On the contrary civic engagement can be a powerful tool in combating this problem. However, the influencing factors that encourage the active participation of young adults yet to be identified. The purpose of this paper is to fill the research gap by creating and validating a research model by including three motives social presence commitment and online offline civic engagement. Design/methodology/approach The study took a quantitative approach to conduct a cross-sectional study. In total, 214 data were collected from the member of a Facebook group of Bangladesh named Foodbank, a restaurant review page through the online questionnaire. After that structural equation modelling techniques have been used to analyse the data, test the model validity and hypothesis. Findings The result shows that both commitment and social presence influence offline and online civic engagement. Excitement motives have a higher effect than information and convenience motive. Besides, 8 out of 10 hypotheses have shown significant results, with only the convenience motive not having any positive influence and effect on social presence and commitment. Practical implications Almost 47.6 out of 158.5 million are young people who are incapable of contributing fully to national development due to a lack of civic engagement. The outcome of this study will be useful for the Government of Bangladesh, as well as for non-governmental organisations and decision-making authorities to form assessments and develop policy on how to engage the young generation in civic activities to achieve further socio-economic development in the country. Originality/value This study contributes to existing literature with newly developed relationships between social presence-civic engagement and commitment-civic engagement. These unique relationships have been empirically tested and resulted insignificant. The study also identifies that it is vital to engage young people more in social works and increase their participation in offline and online activities.
Civic Engagement through Restaurant Review Page in Facebook
(International Journal of Ethics and Systems, Emerald Publishing, 2021-04-03) Satter, A.K.M. Zaidi; Mahmud, Arif; Rahman, Ashikur; Mahmud, Imran; Akter, Rozina
Purpose Existing literature affirms that almost half of the young generation has remained unemployed worldwide. On the contrary civic engagement can be a powerful tool in combating this problem. However, the influencing factors that encourage the active participation of young adults yet to be identified. The purpose of this paper is to fill the research gap by creating and validating a research model by including three motives social presence commitment and online offline civic engagement. Design/methodology/approach The study took a quantitative approach to conduct a cross-sectional study. In total, 214 data were collected from the member of a Facebook group of Bangladesh named Foodbank, a restaurant review page through the online questionnaire. After that structural equation modelling techniques have been used to analyse the data, test the model validity and hypothesis. Findings The result shows that both commitment and social presence influence offline and online civic engagement. Excitement motives have a higher effect than information and convenience motive. Besides, 8 out of 10 hypotheses have shown significant results, with only the convenience motive not having any positive influence and effect on social presence and commitment. Practical implications Almost 47.6 out of 158.5 million are young people who are incapable of contributing fully to national development due to a lack of civic engagement. The outcome of this study will be useful for the Government of Bangladesh, as well as for non-governmental organisations and decision-making authorities to form assessments and develop policy on how to engage the young generation in civic activities to achieve further socio-economic development in the country. Originality/value This study contributes to existing literature with newly developed relationships between social presence-civic engagement and commitment-civic engagement. These unique relationships have been empirically tested and resulted insignificant. The study also identifies that it is vital to engage young people more in social works and increase their participation in offline and online activities.
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
Dark sIde of Mobile Phone Technology: Assessing The Impact of Self-Phubbing and Partner Phubbing On Life Satisfaction
(Informing Science Institute, 2023-12-20) Mahmud, Imran; Jahan, Nusrat; Begum, Afsana; Masud, Adiba
Aim/Purpose The study aims to explore the attributes of self-phubbing and partner- phubbing, as well as their impact on marital relationship satisfaction and the quality of communication. Furthermore, it aims to comprehend how these characteristics could impact an individual’s total level of life satisfaction. Background The study aims to establish a clear association between specific mobile phone usage behaviors and their subsequent impact on relationship satisfaction and the quality of communication. This study investigates the effects of two types of behaviors on interpersonal relationships: self-phubbing, which refers to an indi- vidual being deeply absorbed in their own mobile phone use, and partner-phub- bing, which refers to witnessing one’s partner being deeply absorbed in a mo- bile device.
Dataset on the Influence of Software Development Agility on Software Firms' Performance in Bangladesh
(Data in Brief, Elsevier, 2019-04) Sadia, Farzana; Mahmud, Imran; Dhar, Eva; Jahan, Nusrat; Hossain, Syeda Sumbul; Satter, A.K.M.Zaidi
The article identifies the relationship among different agile software development approaches such as response extensiveness, response efficiency, team autonomy, team diversity, and software functionality that software teams face difficult challenges in associating and achieving the right balance between the two agility dimensions. This research strategy, in terms of quantity, is descriptive and correlational. Statistical analysis of the data was carried out, using SmartPLS 3.0. Statistical population, consist of employees of software industries in Bangladesh, who were engaged in 2017 and their total number is about 100 people. The data show that the response extensiveness, response efficiency, team autonomy, team diversity, and software functionality have impact on software development agility and software development performance.
DeepQSP: Identification of Quorum Sensing Peptides Through Neural Network Model
(Elsevier, 2024-09-13) Rahman, Md. Ashikur; Ali, Md. Mamun; Ahmed, Kawsar; Mahmud, Imran; Bui, Francis M.; Chen, Li; Kumar, Santosh; Moni, Mohammad Ali
Quorum Sensing Peptides (QSP) are small molecules crucial for microbial communication, enabling bacterial populations to coordinate behaviors such as biofilm formation and virulence. The identification of QSP is vital for understanding these biological processes. While existing clinical and lab-based methods are available, they can be costly and time-consuming. This study introduces Deep QSP, a novel technique for QSP identification, which combines Latent Semantic Analysis (LSA), a word embedding feature extraction method, with classical amino acid-based extraction Pseudo Amino Acid Composition (PAAC), and a convolutional neural network (CNN) classifier. The DeepQSP model was evaluated using a dataset of 440 peptide sequences, achieving impressive performance metrics: 0.9697 accuracy, 0.9655 sensitivity, 0.9730 specificity, and a Matthews correlation coefficient (MCC) of 0.9385. The LSA combined with PAAC improves peptide sequence representation, while the CNN effectively captures complex patterns, leading to accurate QSP identification. These quantified results demonstrate the effectiveness of the Deep QSP method, offering a powerful tool for advancing the study of microbial interaction and quorum sensing. The enhanced identification of QSPs is critical for microbiology and bioengineering, aiding in the understanding of cell-to-cell communication in microorganisms.
Digital Game-based Education: A Meta Analysis
(International Conference of Inclusive Innovation and Innovative Management (ICIIIM 2015), 2015-11-26) Bhuiyan, Touhid; Mahmud, Imran
In this paper a quantitative meta-analysis by systematically review is designed and the impacts of digital games in education is analyzed. A major purpose of this literature review and meta-analysis are to define/identify policy and practice based on existing studies and also their impact compare to traditional lecture. This meta-analysis is performed to understand how games based learning can affect students’ knowledge and skills. Several provoking conclusions in the terms of “Game-based education” have been derived. This paper investigates the effectiveness of computer-based instructional games which have been emerged from a comparative analysis with 30 instructional gaming studies and are discussed with the support of appropriate tables. This meta-analysis can influence academicians as an earlier step to rethink delivery approach in traditional pedagogy and identification of benefits of digital game in education.
Does Usability Matter? An Analysis of the Impact of Usability on Technology Acceptance in ERP Settings
(Informing Science, 2016) Scholtz, Brenda; Mahmud, Imran; Ramayah, T.
Though the field of management information systems, as a sector and a discipline, is the inventor of many guidelines and models, it appears to be a slow runner on practical implications of interface usability. This usability can influence end users’ attitude and behavior to use IT. The purpose of this paper was to examine the interface usability of a popular Enterprise Resource Planning (ERP) software system, SAP, and to identify related issues and implications to the Technology Acceptance Model (TAM). A survey was conducted of 112 SAP ERP users from an organization in the heavy metal industry in Bangladesh. The partial least squares technique was used to analyze the survey data. The survey findings empirically confirmed that interface usability has a significant impact on users’ perceptions of usefulness and ease of use which ultimately affects attitudes and intention to use the ERP software. The research model extends the TAM by incorporating three criteria of interface usability. It is the first known study to investigate usability criteria as an extension of TAM. Full Text Link: https://www.informingscience.org/Publications/3591
E-waste Recycling Intention Paradigm of Small and Medium Electronics Store Managers in Bangladesh
(Waste Management and Research, 2020-05-04) Mahmud, Imran; Sultana, Sadia; Rahman, Ashikur
Each year Bangladesh produces around 400,000 metric tonnes of e-waste. E-waste accumulation is expected to increase by 20% annually. In order to facilitate e-waste recycling, it is crucial to identify the factors. In this study, building on the stimulus-organism-response framework, we develop a research model to explore the effect of information publicity, ascription of responsibility and convenience of recycling on the recycling attitude, subjective norm, personal norm and perceived behaviour control which lead to recycling intention. Data were gathered from 127 small and medium electronics store managers. The structural equation modelling technique was used to test the paths. The result suggests a significant influence of the element of stimulus (S) on the element of organism (O). The relationship between the element of organism (O) and the element of response (R) is partial. This paper contributes to the body of work dedicated to helping us better understand the recycling behaviour from the stimulus-organism-response perspective. From the viewpoint of practice, this research sheds light on some of the challenges that the implementer might face when making strategy and policy for e-waste management in Bangladesh.