Browsing by Author "Sarker, Kaushik"

Now showing 1 - 11 of 11

An Investigation and Evaluation of N-Gram, TF-IDF and Ensemble Methods in Sentiment Classification
(Lecture Notes of the Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering, LNICST, 2020) Rahman, Sheikh Shah Mohammad Motiur; Biplob, Khalid Been Md. Badruzzaman; Rahman, Md. Habibur; Sarker, Kaushik; Islam, Takia
In the area of sentiment analysis and classification, the performance of the classification tasks can be varied based on the usage of text vectorization and feature extraction methods. This paper represents a detailed investigation and analysis of the impact on feature extraction methods to attain the highest classification accuracy of the sentiment from user reviews. Unigram, Bigram and Trigram are applied as n-gram vectorization models with TF-IDF features extraction method individually. Accuracy, misclassification rate, Receiver Operating Characteristics (ROC) and recall-precision are used in this study to evaluate which are counted as the most important performance measurement parameters in machine learning based approaches. Parameters are measured by the output obtained from Bagged Decision Tree (BDT), Random Forest (RF), Ada Boost (ADA), Gradient Boost (GB) and Extra Tree (ET). The outcomes of this study is to find out the best fitted combination of term frequency–inverse document frequency (TF-IDF) and n-grams for different data size.
Chirped Large Mode Area Photonic Crystal Modal Fibers and Its Resonance Modes Based on Finite Element Technique
(Journal of Optical Communications, 2019-07-23) Amiri, Is; Rashed, Ahmed Nabih Zaki; Sarker, Kaushik; Paul, Bikash Kumar; Ahmed, Kawsar
The study has outlined the finite element technique used for the fiber modal analysis of photonic crystal fiber (PCF) structure with a hexagonal/circular air holes arrangement on the cladding of pure fiber-optic silica. The used fibers that are namely highly nonlinear fiber (HNLF), single model silica fiber (SMSF) are combined with PCFs with different dopants concentration. Leakage loss, number of guided resonant modes, fiber birefringence, effective refractive index and cross-section areas and nonlinear coefficient parameters are measured inaccurate estimation. The optimum resonant guided modes are also estimated for different lattice infrastructure in circular PCFs.
Comparing the performance of different ultrasonic images enhancement for speckle noise reduction in ultrasound images using techniques: a preference study
(SPIE Digital Library, 2017-06-19) Rana, Md. Shohel; Sarker, Kaushik; Bhuiyan, Touhid; Hassan, Md. Maruf
Diagnostic ultrasound (US) is an important tool in today's sophisticated medical diagnostics. Nearly every medical discipline benefits itself from this relatively inexpensive method that provides a view of the inner organs of the human body without exposing the patient to any harmful radiations. Medical diagnostic images are usually corrupted by noise during their acquisition and most of the noise is speckle noise. To solve this problem, instead of using adaptive filters which are widely used, No-Local Means based filters have been used to de-noise the images. Ultrasound images of four organs such as Abdomen, Ortho, Liver, Kidney, Brest and Prostrate of a Human body have been used and applied comparative analysis study to find out the output. These images were taken from Siemens SONOLINE G60 S System and the output was compared by matrices like SNR, RMSE, PSNR IMGQ and SSIM. The significance and compared results were shown in a tabular format. Full Text Link: https://doi.org/10.1117/12.2280277
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.
Embedded Subscriber Identity Module with Context Switching
(Communications in Computer and Information Science, Springer, 2019-11-13) Sarker, Kaushik; Islam, K. M. Muzahidul
Telecommunications technology user wants quality channels including security, customization, personalization and autonomy. These requirements are encouraging customers to use multiple identity modules. Use of multiple modules brings the necessity of caring multiple cellular phones or multiple cellular units in a mobile device. Moreover, the switching between the identity modules is physical and less secure with the possibility of an identity module cloning or loss of module in case of theft. Several research works have been found in the area of embedded Subscriber Identity Module (eSIM) and Virtual Subscriber Identity module (VSIM). However, the limitations of both eSIM and VSIM have given the authors of this paper a scope to study and come up with a solution by proposing a new model. In the proposed model authors have considered the benefits of eSIM and VSIM together and reducing the limitations such as switching between the modules, parallel activation of the modules, module cloning etc. At the end of the paper, authors have compared ten such limitations, known as features with the existing models and presented a graphical simulation of the proposed model with the proposed methodology of context switching between embedded subscriber identity modules.
Extremely Low Loss Optical Waveguide for Terahertz Pulse Guidance
(Results in Physics, Elsevier, 2019-12) Paul, Bikash Kumar; Abdulrazak, Lway Faisal; Bhuiyan, Touhid; Sarker, Kaushik; Hassan, Md. Maruf; Shariful, S.; Ahmed, Kawsar
This study proposed a dielectric terahertz (THz) D-shape core based photonic crystal fiber (PCF) with very low level of effective material loss (EML) for efficient THz pulse propagation. The modal parameters of the proposed fiber have been rigorously computed using finite element method (FEM) by considering the absorption boundary condition (ABC) using perfectly matched layer (PML). Investigation results of the fiber exhibit a low material absorption loss of 0.027 cm−1 and zero flatted dispersion within the 0.85–1.25 THz range. Additionally, large number of crucial features of the fiber have been evaluated. It reduces the bulk absorption loss of ~87% at the same time offers single mode operation. The proposed fiber can be easily fabricated by using stack and draw or sol-gel technique. Due to excellent optical guiding properties, it can be a potential prospect in THz sensing, communication and imaging applications.
Medicine Prediction Based on Doctor’s Degree
(Daffodil International University, 2022-05-05) Arman, Md Shohel; Sarker, Kaushik; Shakir, Asif Khan; Hossain, Shah Fahad; Hasan, Afia
The effective use of information mining in profoundly unmistakable fields like e - business, promoting and retail has prompted its application in different enterprises. There is an absence of powerful investigation devices to find concealed connections and patterns in information. This examination paper expects to give a review of ebb and flow systems of learning reve lation in databases utilizing information mining strategies that are being used in today’s therapeutic research especially in medicine prediction. Correlation, Chi - square and Euclidean distance feature selections are used to select features and showing the comparison of the result between K - Nearest neighbors, Naïve Bayes, decision tree, artificial neural network . The result uncovers that decision tree beats and sometime Bayesian grouping is having comparative precision as of choice tree. The analysis of per formance can be done in such as doctor’s degrees may vary the diseases medicine.
Prediction Model for Prevalence of Type-2 Diabetes Complications with ANN Approach Combining with K-Fold Cross Validation and K-Means Clustering
(Lecture Notes in Networks and Systems, Springer, 2018-12-06) Munna, Md. Tahsir Ahmed; Alam, Mirza Mohtashim; Allayear, Shaikh Muhammad; Sarker, Kaushik; Ara, Sheikh Joly Ferdaus
In today’s era, most of the people are suffering with chronic diseases because of their lifestyle, food habits and reduction in physical activities. Diabetes is one of the most common chronic diseases which has affected to the people of all ages. Diabetes complication arises in human body due to increase of blood glucose (sugar) level than the normal level. Type-2 diabetes is considered as one of the most prevalent endocrine disorders. In this circumstance, we have tried to apply Machine learning algorithm to create the statistical prediction based model that people having diabetes can be aware of their prevalence. The aim of this paper is to detect the prevalence of diabetes relevant complications among patients with Type-2 diabetes mellitus. The processing and statistical analysis we used are Scikit-Learn, and Pandas for Python. We also have used unsupervised Machine Learning approaches known as Artificial Neural Network (ANN) and K-means Clustering for developing classification system based prediction model to judge Type-2 diabetes mellitus chronic diseases.
Prediction Model for Prevalence of Type-2 Diabetes Complications with ANN Approach Combining with K-Fold Cross Validation and K-Means Clustering
(Scopus, 2020) Munna, Md.Tahsir Ahmed; Alam, Mirza Mohtashim; Allayear, Shaikh Muhammad; Sarker, Kaushik; Ara, Sheikh Joly Ferdaus
In today’s era, most of the people are suffering with chronic diseases because of their lifestyle, food habits and reduction in physical activities. Diabetes is one of the most common chronic diseases which has affected to the people of all ages. Diabetes complication arises in human body due to increase of blood glucose (sugar) level than the normal level. Type-2 diabetes is considered as one of the most prevalent endocrine disorders. In this circumstance, we have tried to apply Machine learning algorithm to create the statistical prediction based model that people having diabetes can be aware of their prevalence. The aim of this paper is to detect the prevalence of diabetes relevant complications among patients with Type-2 diabetes mellitus. The processing and statistical analysis we used are Scikit-Learn, and Pandas for Python. We also have used unsupervised Machine Learning approaches known as Artificial Neural Network (ANN) and K-means Clustering for developing classification system based prediction model to judge Type-2 diabetes mellitus chronic diseases.
Prediction Model for Prevalence of Type-2 Diabetes Complications with ANN Approach Combining with K-Fold Cross Validation and K-Means Clustering
(Springer Nature, 2018-12-06) Munna, Md Tahsir Ahmed; Alam, Mirza Mohtashim; Allayear, Shaikh Muhammad; Sarker, Kaushik; Ara, Sheikh Joly Ferdaus
In today’s era, most of the people are suffering with chronic diseases because of their lifestyle, food habits and reduction in physical activities. Diabetes; is one of the most common chronic diseases which is happened to the people of all ages. Diabetes complication arises in human body due to increase of blood glucose (sugar) level than the normal level. Type-2 diabetes is considered as one of the most prevalent endocrine disorders. Type-2 diabetes is considered one of the most prevalent endocrine disorders. In this circumstance, we have tried to apply Machine learning algorithm to create the statistical prediction based model that people having diabetes can aware of their prevalence. The aim of this paper is to detect the prevalence of diabetes relevant complications among patients with type-2 diabetes mellitus. The processing and statistical analysis we used Scikit-Learn, Pandas for Python. We also have used unsupervised Machine Learning approaches known as Artificial Neural Network (ANN) and K-means Clustering for developing classification system based prediction model to judge type-2 diabetes mellitus chronic diseases.
Supervised Ensemble Machine Learning Aided Performance Evaluation of Sentiment Classification
(IOP Science, 2018-07) Rahman, Sheikh Shah Mohammad Motiur; Rahman, Md. Habibur; Sarker, Kaushik; Rahman, Md. Samadur; Ahsan, Nazmul; Sarker, M. Mesbahuddin
Text vectorization, features extraction and machine learning algorithms play a vital role to the field of sentiment classification. Accuracy of sentiment classification varies depending on various machine learning approaches, vectorization models and features extraction methods. This paper represents multiple ways of evaluations with the necessary steps needed to achieve highest accuracy for classifying the sentiment of reviews. We apply two n-gram vectorization models - Unigram and Bigram individually. Later on, we also apply features extraction method TF-IDF with Unigram and Bigram respectively. Five ensemble machine learning algorithms namely Random Forest (RF), Extra Tree (ET), Bagging Classifier (BC), Ada Boost (ADA) and Gradient Boost (GB) are used here. The key findings in this study is to determine which combination of vectorization models (Bigram, Unigram) along with feature extraction method (TF-IDF) and ensemble classifier gives the better performance of sentiment classification. Full Text Link: https://doi.org/10.1088/1742-6596/1060/1/012036