Scholarly Indexed Publications

Permanent URI for this collectionhttps://hdl.handle.net/10361/28605

Browse

Now showing 1 - 20 of 228

Metadata Only
Player's performance prediction in ODI cricket using machine learning algorithms
(Institute of Electrical and Electronics Engineers Inc., 2018-07-02) Anik, Aminul Islam; Yeaser, Sakif; Imam Hossain, A.G.M.; Chakrabarty, Amitabha; Department of Computer Science and Engineering
This paper presents a method that is aimed towards predicting a cricket player's upcoming match performance by implementing machine learning algorithms. The proposed model consists of statistical data of players of Bangladesh national cricket team which has been collected from trusted sports websites, feature selection algorithms such as recursive feature elimination and univariate selection and machine learning algorithms such as linear regression, support vector machine with linear and polynomial kernel. To implement the proposed model, the accumulated statistical data is processed into numerical value in order to implement those in the algorithms. Furthermore, aforementioned feature selection algorithms are applied for extracting the attributes that are more related to the output feature. Additionally, the machine learning algorithms are used to predict runs scored by a batsman and runs considered by a bowler in the upcoming match. The experimental setup demonstrates that the model gives up to 91.5% accuracy for batsman Tamim and up to 75.3% accuracy for bowler Mahmudullah whereas prediction accuracy for other players are also up to the mark. Therefore, this will help in calculating player's future performance and thus will ensure better team selection for forthcoming cricket matches.
Metadata Only
Current-voltage characteristics of ballistic schottky barrier GNRFET and CNTFET: effect of gate oxide thickness
(Institute of Electrical and Electronics Engineers Inc., 2015-07-01) Ziauddin Ahmed, Sheikh; Shawkat, Mashiyat Sumaiya; Chowdhury, Md. Iramul Hoque; Mominuzzaman, Sharif Mohammad; Department of Electrical and Electronic Engineering
With the advancement of silicon technology having culminated to a point where the scaling limitations of silicon transistors have manifested themselves as being unavoidable, experimentation with Graphene Nanoribbon and Carbon Nanotube field effect transistors has become of the utmost importance. In this paper the effect of gate oxide thickness on the performances of ballistic schottky barrier Graphene Nanoribbon field effect transistor (GNRFET) and ballistic schottky barrier Carbon Nanotube field effect transistor (CNTFET) is studied. A comparative analysis is also done on the two kinds of FETs based on the effect of gate oxide thickness. It has been observed that lowering the gate oxide thickness increases the on-state drain current in both the FETs but the effect is larger in CNTFETs. In this paper a graph is plotted calculating the on-off current ratios of the two transistors, further differentiating their performances with various gate oxide thicknesses.
Metadata Only
Subjectivity analysis of movie reviews using machine learning to perform sentence level classification
(Institute of Electrical and Electronics Engineers Inc., 2018-07-02) Islam, Md. Tanzim; Sazzad, Md. Asad Uzzaman; Ahmed, Erfan; Azad, Muhitun; Islam, Samiul; Department of Computer Science and Engineering
This research conducts a short review, comparative study and performs sentence level subjectivity analysis using three different machine learning algorithms for movie reviews. Support Vector Machine (SVM) and Naïve Bayes Model have been used for subjective/objective classification using the same data-set (rotten tomato movie reviews) to analyze the result as well as to find the best outcome. Multilayer perceptron (MLP) has been implemented later to the feasibility of it's in differentiating subjective and objective notions. In sentence level classification, using stop words produce a better result than not using stops words. A comparative analysis has also been included to support this claim.
Metadata Only
Current-voltage characteristics of ballistic schottky barrier GNRFET and CNTFET: effect of relative dielectric constant
(Institute of Electrical and Electronics Engineers Inc., 2015-07-01) Ziauddin Ahmed, Sheikh; Shawkat, Mashiyat Sumaiya; Chowdhury, Md. Iramul Hoque; Mominuzzaman, Sharif Mohammad; Department of Electrical and Electronic Engineering
Graphene Nanoribbon (GNR) and Carbon Nanotube (CNT) are currently being considered as two of the most promising options to replace silicon technology. Silicon technology is faced with scaling limits and other material issues which hinder the development of transistor technology. In this paper, the effect of relative dielectric constant on the performances of ballistic schottky barrier Graphene Nanoribbon field-effect transistor (GNRFET) and Carbon Nanotube field-effect transistor (CNTFET) is studied and a comparative analysis between the two transistors is provided. It has been observed that using a gate material with higher relative dielectric constant leads to a higher on-state drain current for both the transistors. However, CNTFET has higher on-state drain current compared to GNRFET. Also in this literature, the on and off-state current ratios of both the transistors are calculated and plotted to further differentiate between the performances of GNRFET and CNTFET.
Metadata Only
Transition in electronic transport properties of graphene nanoribbon due to the adsorption of atoms and molecules
(Institute of Electrical and Electronics Engineers Inc., 2015-07-01) Shakil, Shifur Rahman; Atanu Kumar, Saha; Hossain Bhuian, Md. Belal; Department of Electrical and Electronic Engineering
The adsorption effects of gas molecules on semiconducting armchair graphene nanoribbon (A-GNR) are studied considering on the significant changes over current-voltage characteristics, device density of states and electrostatic difference potential. The orientation of gas molecules and adsorption sites play an important role on charge transfer between the graphene surface and the molecules. The charge transfer mechanism is discussed in the light of molecular orbital theory. We have used three gas molecules-CO2, NH3 and NO as adsorbates and chose such orientation that adsorbates can act only as donor. Finally, we have made a conclusion that semiconducting A-GNR shows metallic behaviour after adsorbing optimum number of adsorbates corresponding to the area of A-GNR. Such semiconducting to metal transition can be used as sensing parameters.
Metadata Only
Vapor adsorption limitation of graphene nanoribbons in quasi conductance increment: a NEGF approach
(Institute of Electrical and Electronics Engineers Inc., 2015-07-01) Shakil, Shifur Rahman; Zohra, Fatema Tuz; Pramanik, Parna; Tushar, Raihanul Islam; Atanu Kumar, Saha; Hossain Bhuian, Md. Belal; Department of Electrical and Electronic Engineering
The adsorption of H2O molecules on Armchair Graphene Nanoribbons (A-GNR) was theoretically studied using Non Equilibrium Green Function (NEGF) formalism to determine Device Density of States (DDOS), Electrostatic Effective Potential (EDP), Conductivity (G) and Current-Voltage (I-V) characteristics. This paper analyzed the performance of semiconducting graphene nanoribbon (N=10), metallic graphene nanoribbon (N=11) and cascade hetero-graphene nanoribbon to consider the effect of H2O adsorption on GNR and conclude that for low voltage application semiconducting A-GNR will be a better choice over metallic A-GNR and cascade A-GNR. The optimum area for adsorbing maximum number of H2O molecules on A-GNR has been studied. It has been observed that the increment of quasi conductance resulting sensing performance for limited number of H2O adsorption. To overcome this problem, a new device model has been proposed.
Metadata Only
Current-voltage characteristics of CNTFET considering non-ballistic conduction: effect of dielectric constant
(Institute of Electrical and Electronics Engineers Inc., 2014-09-23) Rouf, Nirjhor Tahmidur; Deep, Ashfaqul Haq; Hassan, Rusafa Binte; Khan, Sabbir Ahmed; Hasan, Mahmudul; Mominuzzaman, Sharif Mohammad; Department of Electrical and Electronic Engineering
Carbon nanotube is being considered as a prospective alternative for the existing Silicon technology in present days. The advancement of Silicon technology, at present, has slowed down considerably because of its various material issues and scaling limitations and so, carbon nanotubes have gained the attention of researchers around the world over. In this literature, the effect of change of dielectric constant of the gate material on the performance of carbon nanotube field effect transistor (CNTFET) with non-ballistic conduction is extensively investigated. It is found in this study that with the value of dielectric constant of the gate material greatly improves the current output of CNTFET considering Non-Ballistic conduction. In addition to this, the obtained results are compared with previously reported CNTFET with ballistic conduction data for a better perception of the non-ballistic conduction with effects. Also, the on and off state current ratios have been calculated and plotted in this work to further contrast the ballistic and non-ballistic conduction of CNTFET.
Metadata Only
Epitaxial growth controlled tailoring of Metal-Insulator (MI) transition properties of rare earth correlated oxides
(Institute of Electrical and Electronics Engineers Inc., 2014-09-23) Iqbal, A.; Khan, S.A.; Rahman, N.U.; Faraz, T.
Strongly correlated electron devices using Metal Insulator Transition (MIT) Oxides are prospective alternatives along the new generation of high speed devices based on novel mechanisms. Taking the advantages of correlated electrons which are capable of forming a variety of electronic phases, MIT Oxides and Phase Change Materials (PCM) are treated as the frontiers of emergent device research. With the prospect of downsizing devices to the nanoscale regime, benefits over conventional semiconductor devices are attained. Aided by recent advances in fabrication technology, considerable improvements have been achieved to tailor the Metal-Insulator (MI) transition properties of MIT Oxides. In this study, the tailoring of MI transition properties for a particular group of MIT Oxides, namely the transition metal perovskite oxides of RNiO3 family are studied on the epitaxial platform. Finally, antiferromagnetism characteristics and anonymous resistivity inherent within those oxides are studied.
Metadata Only
Hybrid email filtering using TF-IDF and BERT enhanced spam detector
(Institute of Electrical and Electronics Engineers Inc., 2025-01-01) Sarker, Sumit Kanti; Rahman, Mohammad Shoaib; Al Prince, Abdullah; Al Mahmud Riaz, Abdul; Sumon, Md Shakhauat Hossan; Sakib, Md. Tauhidur Rahman; Sohanoor, Sifat; Talha, Md Abu; Department of Computer Science and Engineering
Email spam detection is a crucial cybersecurity challenge requiring accurate and efficient filtering methods. This study introduces the TF-IDF and BERT Enhanced Spam Detector (TBESD), a hybrid deep learning model that integrates statistical and contextual text analysis for robust spam classification. TF-IDF captures essential n-gram patterns, while BERT embeddings extract deep semantic features. These feature vectors are combined and processed through a multilayer perceptron (MLP) with dropout regularization and ReLU activation for optimal classification. The methodology includes text preprocessing (normalization, stopword removal, tokenization), feature extraction (TF-IDF, BERT embeddings), and model training using the Adam optimizer with binary cross-entropy loss. Evaluated on a publicly available dataset, TBESD achieved an impressive accuracy of 98.88%, precision of 98.44%, recall of 99.44%, F1-score of 98.94%, and an AUC score of 0.9990. This study demonstrates the superiority of hybrid models over conventional methods, highlighting the role of advanced NLP techniques in cybersecurity.
Metadata Only
Detecting jute plant disease using image processing and machine learning
(Institute of Electrical and Electronics Engineers Inc., 2017-03-06) Reza, Zarreen Naowal; Nuzhat, Faiza; Mahsa, Nuzhat Ashraf; Ali, Md. Haider; Department of Computer Science and Engineering
Detecting stem diseases of plants by image analysis are still in an inchoate state in the research field. This research has been conducted on detecting the stem diseases of jute plants which is one of the most important cash crops in some of the Asian countries. An automated system based on an Android application has been implemented to take pictures of the disease affected stems of jute plants and send them to the dedicated server for assaying. On the server side, the affected portion from the image will be segmented using customized thresholding formula based on hue-based segmentation. The consequential feature values will be extracted from the segmented portion for texture analysis using color co-occurrence methodology. The extracted values will be compared with the sample values stored in the pre-defined database which will lead the disease to be identified and classified using Multi-SVM classifier. At the final step, the classification result along with the necessary control measures will be sent back to the farmer within three seconds through the application on their phone.
Metadata Only
Covid-19 challenges and responses in Bangladesh
(National Bureau of Asian Research, 2022-01-01) Alam, Tasnia
Metadata Only
Cutting-edge deep learning architectures for detecting and classifying phishing URLs
(Institute of Electrical and Electronics Engineers Inc., 2025-01-01) Mukti, Zinia Sultana; Sultana, Taniya; Sumaia, Subrin; Sultana, Abida; Department of Computer Science and Engineering
Phishing is an increasingly prevalent cyberthreat that exploits technological and psychological vulnerabilities to result in substantial money losses and data exposures. Being alert is crucial since phishing can result in identity theft, financial loss, and security breaches. This study intends to develop and assess sophisticated deep learning models for automated phishing URL classification in order to address the swift evolution of phishing tactics and offer a reliable substitute for conventional detection techniques. The proposed model utilizes state-of-the-art architectures that combine character- and word-level embeddings with a set of hand-crafted features specifically designed to capture subtle indicators of phishing. A comparative analysis of Convolutional Neural Networks (CNN), Deep Neural Networks (DNN), and Gated Recurrent Units (GRU) reveals that CNN achieved a 96.31, DNN reached the highest accuracy of 97.81 addition of handcrafted features alongside character- and wordlevel embeddings significantly improved model performance, consistently surpassing conventional methods in all models tested. The integrating features not only enhances detection accuracy but also offers a resilient and adaptable solution to evolving phishing threats.
Open Access
Barriers and facilitators to women's leadership in savings associations in Uganda
(Policy Press, 2023-06-01) Torrance, Robert James; Sikoti, Doreen; Sulaiman, Munshi; Mulyampiti, Tabitha; Nabulumba, Christine; Namisango, Eve; Mutebi, Ronald; Cislaghi, Beniamino
The Village Savings and Loan Association (VSLA) model is currently being employed in Uganda for deepening financial inclusion and poverty reduction. Despite its focus on women's empowerment, concerns have arisen of an under-representation of women on VSLA leadership committees. Human rights-based, economic, and social justice arguments support active participation of women on VSLA leadership committees. The study sought to identify, explicate and characterise the barriers and facilitators to women in VSLA leadership. An exploratory study design using qualitative methods was selected to address the research objectives. Forty-nine focus group discussions were undertaken, featuring both VSLA members and non-members. VSLAs for inclusion in the study were randomly selected from within four regions of Uganda, stratified by: mature (>2 years old) versus new (<2 years old). The study exposed a diverse array of barriers and facilitators to women in VSLA leadership positions, revealing the influence of individual, material, institutional and social factors, in addition to social norms and gender characteristics, on women in VSLA leadership. The findings revealed that the design of interventions to achieve fair representation of women in leadership positions should be informed by an understanding of the different types, relative strengths, support for/against, and intersectionality of the factors impacting women in VSLA leadership. © Policy Press 2023.
Metadata Only
A comparative analysis between light dependent and ultrasonic method of gesture recognition
(Institute of Electrical and Electronics Engineers Inc., 2017-03-06) Uzzaman N.; Hossain S.; Rashid, Rezaur; Hossain A.; Department of Computer Science and Engineering
In this paper, two methods of gesture recognition are discussed and a comparison between these methods is carried out. These are light dependent and ultrasonic method of gesture recognition. The light dependent method was developed in an early study, where a control panel detected some predefined gesture commands and a robotic arm was controlled with those commands. In this new study the same gesture commands are used to control the robotic arm. But this time they are determined by both light dependent and ultrasonic method. All the procedure and arrangements are kept unchanged so that a comparative performance test for these two methods can be carried out. Experimental results and cost analysis shows that the light dependent method is more convenient for indoor application of the system. Whereas, for outdoor application, the ultrasonic method is preferable.
Open Access
A comparative study of space and time fractional KdV equation through analytical approach with nonlinear auxiliary equation
(Horizon Research Publishing, 2020-01-01) Naher, Hasibun; Shafia, Humayra; Ali, Md. Emran; Paul, Gour Chandra; Department of Mathematics and Natural Sciences
In this article, the nonlinear partial fractional differential equation, namely the KdV equation is renewed with the help of modified Riemann-Liouville fractional derivative. The equation is transformed into the nonlinear ordinary differential equation by using the fractional complex transformation. The goal of this paper is to construct new analytical solutions of the space and time fractional nonlinear KdV equation through the extended (G'/ G)-expansion method. The work produces abundant exact solutions in terms of hyperbolic, trigonometric, rational, exponential, and complex forms, which are new and more general than existing results in literature. The newly generated solutions show that the executed method is a well-organized and competent mathematical tool to investigate a class of nonlinear evolution fractional order equations. © 2020 by authors, all rights reserved.
Metadata Only
A comparative analysis of deep learning and hybrid models to diagnose multi-class skin cancer
(Institute of Electrical and Electronics Engineers Inc., 2023-01-01) Nawrin, Ishrat Nur; Trina, Tonusree Talukder; Rahman, Rafeed; Rasel, Annajiat Alim; Department of Computer Science and Engineering
Skin cancer is one of the most lethal and increasingly prevalent cancers in the world. Skin cancer develops when the epidermal (top layer of skin) cells divide abnormally, causing it to spread to other regions of the human body. Skin cancer exists in seven different varieties. The presence of malignant epidermal cells determines the type of skin cancer. Dermoscopy, spectroscopy, and imaging tests are primarily utilized to identify the malignancy. These procedures are expensive and prolonged. It may result in unfavorable effects such as bleeding, bruising, and infection as well. The narrow variances in multi class cancer pictures escalate the complexity of classification. Dermatologists confront challenges in the categorization of cancer types from images. Deep learning has resulted in a dramatic leap in disease identification. Deep learning models are capable of categorizing skin cancer more precisely than dermatologists. Several studies focused on pretrained and hybrid models for categorizing the classes of skin cancer. In contrast to binary classification, the multi-class classification of skin cancer yielded a mediocre result for both deep learning and dermatologists. The proposed study employs varieties of deep learning and hybrid models to examine the performance of each model in categorizing the classes of cancer. The proposed CNN-LSTM-SVM hybrid model obtained the highest result compared to other models with 88.24% accuracy. To illustrate the overall comparison, the models have been depicted in an evaluation matrix table.
Open Access
A novel modified SFTA approach for feature extraction
(Institute of Electrical and Electronics Engineers Inc., 2017-03-06) Hasan, Md. Junayed; Uddin, Jia; Pinku, Subroto Nag; Department of Computer Science and Engineering
To increase the efficiency of conventional Segmentation Based Fractal Texture Analysis (SFTA), we propose a new approach on SFTA algorithm. We use an optimum multilevel thresholding hybrid method of Genetic Algorithm (GA) and Particle Swarm Optimization (PSO), called HGAPSO with the optimization technique for classification based on grey level range to get more accurate output. Experimental results show that proposed approach exhibits average 2% higher classification accuracy than conventional SFTA for our tested dataset.
Open Access
Role of customary forest recognition and social trust in community-based biodiversity conservation in Bungo, Jambi, Indonesia
(Society for Indonesian Biodiversity, 2025-01-01) Ridwan; Maryam, Siti; Epriadi, Dedi; Rizki; Alam, Mahmudul; Sari, Susi Puspita; Kirin, Arwansyah; BRAC Business School
Forests, including those governed by customary practices, are critical ecosystems that help maintain global climate balance. This study explores how customary forest recognition and social beliefs shape community behavior in protecting biodiversity in the Datuk Sinaro Putih customary forest in Jambi, Indonesia. Data were collected using a descriptive qualitative design through 17 semi-structured interviews with customary leaders, village officials, members of the Customary Forest Management Institution (LPHA), youth, community leaders, and non-governmental organizations from February to October 2024. Data were refined through participant observation and document analysis. The findings indicate that customary forest recognition and social beliefs are positively correlated with conservation behavior. Seloko Adat, the indigenous oral tradition of Datuk Sinaro Putih local community, serves as a symbolic and behavioral system, preserving traditional ecological knowledge, upholding customary law, and fostering intergenerational responsibility. This study highlights how the recognition of customary forests and social trust not only discourages extractive behavior through customary sanctions but al so embeds conservation values into people's daily lives. These findings suggest that forestry policies should recognize and strengthen the functions of customary institutions through legal support, institutional strength, and sustainable financing schemes. Policies that incorporate customary principles, such as ecological prohibitions and collective farming practices, can enhance the legitimacy and success of conservation while safeguarding ecosystems. Future research should include a more comprehensive multi-level governance model that integrates local wisdom with national policies and the global biodiversity agenda, and analysis of economic, spiritual, and cosmological dimensions, as well as the relationships between customary institutions and external actors. © 2025, Society for Indonesian Biodiversity. All rights reserved.
Open Access
A review of the energy-employment Nexus in Bangladesh: Rural-urban electrification and sectoral occupation patterns
(River Publishers, 2022-01-01) Rahman, Syed Mahbubur; Kirshner, Joshua; Groh, Sebastian; Rahman, Syed Mustafizur; BRAC Business School
Energy access, more specifically electricity access, is one of the prerequisites to achieve the Sustainable Development Goals. Around one-tenth of the global population remained without access to electricity in 2019. Bangladesh has shown substantial progress in electricity access. This paper aims at reviewing the association between electricity access and employment in the core three sectors: agriculture, industry, and service. The study finds that rural access to electricity is steeper in progress than that of urban areas; and overall access to electricity is increasing, while the urban-rural disparity in energy access is decreasing over time. The proportion of female employment in agriculture has reduced by one third during 1991-2018. Agricultural employment, in general, decreases with an increase in electricity access, which is counterintuitive. The government should consider the issue to keep the agricultural employment stable to ensure food security and the sector's contribution to gross domestic product. © 2022 River Publishers
Open Access
In vitro screening for phytochemicals and antioxidant activities of Syngonium podophyllum L.: An incredible therapeutic plant
(Oriental Scientific Publishing Company, 2017-01-01) Hossain, Sarwar; Uddin, Md.Sahab; Kabir, Tanvir; Begum, Marium; Koushal, Paras; Herrera-Calderon, Oscar; Akter, Raushanara; Asaduzzaman M.D.; Abdel-Daim, Mohamed M.; School of Pharmacy
Searching for natural curative agents with a superior safety profile for the management of oxidative stress linked disorders are continuing owing to multiple unwanted effects linked with synthetic antioxidants. Medicinal plants are still considered as one of the imperative cradles of modern medicine. Therefore the drive of this study was to analyse the phytochemical constituents and antioxidant activities of the Syngonium podophyllum (S. podophyllum) L. leaves. The fresh leaves of S. podophyllum L. were extracted with methanol (CME) afterward fractionation using n-hexane (NHF), chloroform (CLF), ethyl acetate (EAF) and aqueous (AQF). The plant extract and its fractions were evaluated for phytochemical content by alkaloids, carbohydrates, saponins, tannins, resins, flavonoids and steroids tests. Total phenolic content (TPC) was also determined. Antioxidant activities were determined by total antioxidant activity (TAA), reducing power activity (RPA) and DPPH (2,2-diphenyl-1-picrylhydrazyl) radical scavenging activity tests. The phytochemical analysis of CME and its fractions of S. podophyllum L. leaves exposed the presence of alkaloids, tannins, resins, flavonoids and steroids. The CME of S. podophyllum L. leaves exhibited the highest amount (10.55 mg of GAE/g of dried extract) of TPC compared to remaining fractions. The highest absorbance, 3.791 nm was found in CME with respect to remaining fractions for TAA test. Among the crude extract and its different fractions, highest absorbance, 3.015 nm was reported for CME in RPA test. In the DPPH radical scavenging test, CME showed the highest scavenging activity (79.89%) with IC50 of 46.55 ìg/mL. This study indorsed that CME has marked antioxidant activities. This plant extract might be a promising source of natural antioxidant which can be effective for the treatment of oxidative stress related disorders.

Browse

Recent Submissions