检索结果-南通市图书馆

Performance Evaluation of Multiple Classifiers for Predicting Fake News

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journal of Computer and Communications 2022年第9期10卷 1-21页

作者： Arzina Tasnim md. Saiduzzaman Mohammad Arafat Rahman Jesmin Akhter abu sayed md. mostafizur rahaman Department of Information Systems Security Bangladesh University of Professionals Dhaka Bangladesh Institute of Information Technology Jahangirnagar University Dhaka Bangladesh Department of Computer Science and Engineering Jahangirnagar University Dhaka Bangladesh

The rise of fake news on social media has had a detrimental effect on society. Numerous performance evaluations on classifiers that can detect fake news have previously been undertaken by researchers in this area. To assess their performance, we used 14 different classifiers in this study. Secondly, we looked at how soft voting and hard voting classifiers performed in a mixture of distinct individual classifiers. Finally, heuristics are used to create 9 models of stacking classifiers. The F1 score, prediction, recall, and accuracy have all been used to assess performance. Models 6 and 7 achieved the best accuracy of 96.13 while having a larger computational complexity. For benchmarking purposes, other individual classifiers are also tested.

关键词： Fake News Machine Learning TF-IDF Classifier Estimator F1 Score Recall Precision Voting Classifiers Stacking Classifier Soft Voting Hard Voting

Towards the Advancement of Cashless Transaction: A Security Analysis of Electronic Payment Systems

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journal of Computer and Communications 2022年第7期10卷 103-129页

作者： Iffath Tanjim Moon Muhammad Shamsuzzaman Muhammad Musfiqur Rahman Mridha abu sayed md. mostafizur rahaman Department of Computer Science and Engineering Begum Rokey University Rangpur Bangladesh Department of Computer Science and Engineering Jahangirnagar University Dhaka Bangladesh

In recent decades, day-to-day lifestyle requires online payments as easy and simple solutions to several financial transactions, which makes the concept of Electronic payment Systems very popular in the growth of a cashless society. In fact, cashless transactions through simple mobile apps are not merely a concept anymore rather are implemented robustly and being used extensively. On the dark side, obvious financial benefits are making these apps vulnerable to being attacked, which can be successful through security breaches. These cybersecurity issues need to be traced out and resolved to make the financial transactions through an app secure and trustworthy. In this paper, several related papers are analyzed to trace out possible cybersecurity issues in the domain of Electronic Transaction System. The objective is to establish sufficient theoretical background to propose methodologies for measuring security issues and also identify the security strength of any FinTech application and provide standard security metrics.

关键词： Fintech Apps Cybersecurity Cashless Transaction Mobile Banking Electronic Payment

Data Classification Using Combination of Five Machine Learning Techniques

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journal of Computer and Communications 2021年第12期9卷 48-62页

作者： md. Habibur Rahman Jesmin Akhter abu sayed md. mostafizur rahaman md. Imd.dul Islam Department of Computer Science and Engineering Jahangirnagar University Dhaka Bangladesh Institute of Information Technology Jahangirnagar University Dhaka Bangladesh

Data clustering plays a vital role in object identification. In real life we mainly use the concept in biometric identification and object detection. In this paper we use Fuzzy Weighted Rules, Fuzzy Inference System (FIS), Fuzzy C-Mean clustering (FCM), Support Vector Machine (SVM) and Artificial Neural Network (ANN) to distinguish three types of Iris data called Iris-Setosa, Iris-Versicolor and Iris-Virginica. Each class in the data table is identified by four-dimensional vector, where vectors are used as the input variable called: Sepal Length (SL), Sepal Width (SW), Petal Length (PL) and Petal Width (PW). The combination of five machine learning methods provides above 98% accuracy of class identification.

关键词： Co-Variance of Fuzzy Rule Objective Function Surface Plot Confusion Matrix Scatterplot and Accuracy of Detection

Determination of Array Gain of Single Hop to Achieve the Performance of a 2-Hop Wireless Link

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journal of Computer and Communications 2018年第7期6卷 84-98页

作者： abu sayed md. mostafizur rahaman Jesmin Akhter md. Rafsan Jani md. Imd.dul Islam Department of Computer Science and Engineering Jahangirnagar University Savar Dhaka Bangladesh Institute of Information Technology Jahangirnagar University Savar Dhaka Bangladesh

In adaptive beamforming system adaptive algorithm of digital filter is applied to update the weighting vector of the antenna elements to get antenna gain along the desired direction and attenuation along the jammer. The objective of the paper is to evaluate the threshold gain of the adaptive beam former along the line of sight (LOS) between the transmitter and the receiver (including jammer suppression) to make the single hop link comparable with 2-hop link. The single hop and 2-hop communication systems are compared in context of symbol error rate (SER) under fading condition theoretically and verified by simulation. Finally we evaluate the numerical value of threshold gain of adaptive beamformer of two antenna elements under Rayleigh and Nakagami-m fading conditions.

关键词： Array Antenna System BER Threshold Array Gain Side Lobe Canceller Path Loss

Traffic Modelling of Low Dense Femtocellular Network for Long Term Evolution

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journal of Computer and Communications 2019年第12期7卷 88-101页

作者： Jesmin Akhter abu sayed md. mostafizur rahaman md. Imd.dul Islam M. R. Amin Institute of Information Technology Jahangirnagar University Dhaka Bangladesh Department of Computer Science and Engineering Jahangirnagar University Dhaka Bangladesh Department of Electronics and Communications Engineering East West University Dhaka Bangladesh

One of the features of Long Term Evolution (LTE) is the deployment of femtocells as the underlain cell of a macrocell without any intervention of frequency planning to offload the traffic. In this paper we used Markov chain to derive the expression of blocking probability for both macro and femtocell in terms of traffic parameters of the network. We developed an analytical model to find the expression of probability of forced termination (FT) using combination of mobility model and probability tree considering low dense femtocellular network. Two different trees were designed: a newly originating call which starts its session in a femtocell and that of in a macro cell. The link parameters of small scale fading of wireless network under Multiple Input Multiple Output (MIMO) are combined with the proposed traffic model to get the probability of FT of a real-life network. A new state transition chain was also developed including its solution for LTE traffic of variable bandwidth (BW) and a comparison was made with Erlang’s traffic model.

关键词： Rayleigh and Nakagami-m Fading MIMO RB Forced Termination Outage Probability

A Critical Analysis of Machine Learning and Deep Learning Methods for Cervical Cancer Screening

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journal of Computer and Communications 2023年第12期11卷 64-85页

作者： Muhtasim Mahmudur Rahman Jakir Khan abu Sale Mohammad mostafizur Rahman Redoanul Haque md. Sumon Ali Department of Computer Science and Engineering Bangladesh Army University of Engineering & Technology Natore Bangladesh

Cervical cancer is a serious public health issue worldwide, and early identification is crucial for better patient outcomes. Recent study has investigated how ML and DL approaches may be used to increase the accuracy of vagina tests. In this piece, we conducted a thorough review of 50 research studies that applied these techniques. Our investigation compared the outcomes to well-known screening techniques and concentrated on the datasets used and performance measurements reported. According to the research, convolutional neural networks and other deep learning approaches have potential for lowering false positives and boosting screening precision. Although several research used small sample sizes or constrained datasets, this raises questions about how applicable the findings are. This paper discusses the advantages and disadvantages of the articles that were chosen, as well as prospective topics for future research, to further the application of ml and dl in cervical cancer screening. The development of cervical cancer screening technologies that are more precise, accessible, and can lead to better public health outcomes is significantly affected by these findings.

关键词： Cervical Cancer Neoplasms Screening Machine Learning Techniques Deep Learning Techniques

Assessment of natural regeneration status and diversity of tree species in the biodiversity conservation areas of Northeastern Bangladesh

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journal of Forestry Research 2011年第4期22卷 551-559页

作者： md. Habibur Rahman md. abu sayed Arfin Khan Bishwajit Roy Most. Jannatul Fardusi Department of Forestry and Environmental Science School of Agricul- ture and Mineral Sciences Shahjalal University of Science and Tech- nology Sylhet-3114 Bangladesh Department of Disturbance Ecology University of Bayreuth Bayreuth Germany

A study was conducted at two-biodivcrsity conservation areas of Northeastern Bangladesh （a part of Sylhet Forest Division） to assess the species composition, diversity and density of natural regeneration of tree species both indigenous and exotic species. Data were collected by stratified random quadrate method during January 2010 to July 2010. Totally 200 circular plots of 2 m〉〈2 m in size had 5 different habitat types of plants namely; forest, roadside, homestead （surrounding forest dwellers house）, fallow land and others （canals, streams and tea gardens side）, which included a total of 55 regenerating species belonging to 28 families. Meliaceae is the dominant family and shows the highest family importance value （26.3）, having six species, followed by Moraeeae （24.24）. Among the five habitat types, forest （43 species） possess the highes number of species, followed by roadside （23 species）. Total 15 exotic species among 9 families and 40 indigenous species with 24 families were recorded. For exotic species, Tectona grandis possess the highest relative density （11.7%） and relative frequency （10.5%）; Senna siamea had highest relative abundance （7.83%）. In case of indigenous species, Chickrassia tabularis possess the highest relative density （4.23%） and relative frequency （4%）; Dipterocarpus turbinatus bad the highest relative abundance （3.92%）. Tectona grandis （29.66） and Chickrassia tabularis （10.8） had the highest IVI for exotic and indigenous species respectively. Different diversity indic＇es such as Shanon-Winner diversity index,＇ species diversity index, species richness index, species evenness index, Simpson index and species dominance index, etc. were applied to quantify definite diversity. The regeneration of species associated with low levelsof disturbance was in the exotic Species. Study suggests that proper protection from human disturbances and scientific management of natural regeneration of two-study forests may lead a rich biodiversity site in the cou

关键词： biodiversity exotic species family importance value indigenous species quantitative characters

Status and ethno-medicinal usage of invasive plants in traditional health care practices: a case study from northeastern Bangladesh

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Journal of Forestry Research 2011年第4期22卷 649-658页

作者： Mohammed abu sayed Arfin Khan Fahmida Sultana md. Habibur Rahman Bishwajit Roy Sawon Istiak Anik Department of Forestry and Environmental Science School of Agriculture and Mineral Sciences Shahjalal University of Science and Technology Sylhet 3114 Bangladesh

Species diversity and ethno-medicinal usage of 39 species of invasive plants were explored in and around two protected areas, Khadimnagar National Park （KNP） and Rema-Kalenga Wildlife Sanctu- ary （RKWS）, of northeastern Bangladesh,. Status of invasive plants were investigated in 60 sample plots with 5 different habitat types, including forest, roadside, homestead, fallow land and others （ponds, canals, water logged areas, agricultural land, etc.）. Data about the usage of invasive plants in traditional health care were collected through interviewing 110 households. Among the five habitat types, fallow land （28 species）.possessed the highest number of species, followed by roadside （25 species）, forest （23 species）, homestead （22 species） and others （13 species）. Based on the survey, invasive plants of study areas were also categorized into five degrees of invasiveness e.g., naturalized, introduced, possibly invasive, moderately invasive and highly invasive. Additionally, there is the linear trend between degrees of invasivaness and use percentage of invasive plants. Total 39 species of invasive plants belonging to 29 families were recorded, which are generally used to treat 37 diseases, ranging from simple headache to highly complicated eye and heart diseases. The majority of the species used by the local inhabitants were herbs （16 species）, followed by some shrubs （11 species）, climbers （5 species）, trees （5 species） and grasses （2 species）. The use percentage of aboveground plant parts were higher （70.58%） than that of underground plant parts （16.18%）.

关键词： diversity usage invasive species medicinal plants ethno-medicine protected areas

Hydrobiological variables as a regulatory factor on the abundance of heterotrophic flagellates in an urban pond

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Journal of Oceanology and Limnology 2019年第3期37卷 1030-1036页

作者： BEGUM Moni JEWEL md.abu sayed HAQUE md.Ayenuddin KHATUN Mst.Samsad Department of Fisheries Faculty of Agriculture University of Rajshahi

The seasonal abundance of flagellates has been monitored over a period of 1 year from December 2013 to November 2014(divided into 4 conjugative seasons namely winter,spring,summer,and autumn)in an experimental pond located in Rajshahi City Corporation area,*** our knowledge,this study is the fi rst to shed light on the occurrence and possible interrelationships among heterotrophic flagellates(HF),bacteria and zooplankton in Bangladesh and the result obtained by this study will be beneficial for similar water ecosystem all over the *** methods were used to determine the prescribed hydrological parameters and zooplankton cell *** HF abundance(14 346.00 cells/mL)was found in the spring and the minimum(5 215.00 cells/mL)occurred in the *** to HF,significantly(Pabundance was found during the winter(782.00±47.62 cells/mL)and the lowest value was found in the autumn(448.00±39.15 cells/mL).Whereas similar to the HF,total bacterial abundance was significantly higher during the spring((2.25±1.05)×10^5 cells/mL)and lower in the summer((0.79±0.06)×10^5 cells/mL).Multivariate analyses(ANOSIM and md.)have shown significant seasonal differences for cell numbers where md. ordination plot and cluster analysis based on similarity in the genera abundance of HF revealed overlapping condition between winter and *** correspondence analysis(CCA)also showed a distinct separation among the genera based on the prevailing hydrological situation and indicated that temperature,pH,BOD5,and NO3^-were the most important environmental variables in determining the observed variation in HF community *** the biological factors,zooplankton showed negative but total bacteria were positively correlated with HF abundance.

关键词： heterotrophic flagellates abundance hydrobiological factors urban pond