Interesting in predictive analytics?
Then research
artificial intelligence,
machine learning,
and deep learning.
AI, ML, Deep Learning, NLP, Big Data & Data Science, Computer vision, Computational Mathematics, Mathematical Model & Algorithm Design, and Statistics.
Many of them worked to classify diabetes patients. With different views and advance work, we did an intelligence diabetes medicine detection technique. we used seven classifier algorithms of machine learning and deep learning. This research will aid the future researcher for developing artificial doctors. This work appeared in Proceedings of International Joint Conference on Computational Intelligence, Springer, Singapore, 2018.
Regression is a procedure to estimate the bond among variables. It is a statistical technique and is used as prediction with the curve fitting in machine learning, data science, economics etc. Linear and Polynomial regression are widely used to fit a curve and forecasting result. In this exploration, we propose two new linear and non-linear regression techniques using the strait of interpolation-extrapolation and Bisection of Numerical Analysis. Basically, interpolation and extrapolation cannot be exerted in regression because of overfitting curve. In our paper, we develop a technique to reduce the curve fitting that will enable the interpolation’s and extrapolation’s scheme to apply in regression. Another procedure is to find out an equation of curve fitting with an optimal way using Bisection method. We also show the graphical presentations and comparison through all the occurring iterations.This work appeared in Algorithms for Intelligent Systems, Springer, 2019.
LSTM is the well-distinguished procedure of chatbot. Nevertheless, if a user makes the line break of sequence then it is rare to inform the right information without affecting the previous knowledge. As a result, in the information desk, LSTM is not a better option for taking steps of the right information. Consequently, mathematical and statistical procedures are prominently good for providing the right knowledge without having back the impact of sequence; but it takes more execution time for providing a reply. The main goal of this paper is to present the optimal chatbot for the lowest execution time and three mathematical and statistical strategies for Bangla Intelligence chatbot in light of information obtained from Noakhali Science and Technology University (NSTU). As the procedures, we have followed cosine similarity, Jaccard similarity, and naive byes classifier. In order to, reduce time and space complexity, we decorated the whole path like as 3-depth tree such as a question, topic, and answer via the assist of NMF and SVD. However, the whole procedures are separated into four parts such as data collection, pre-processing them, reducing time and space complexity, and performing algorithms. Besides, evaluating & compare among three strategies where the best performer accuracy is 96.22%.
Smart Innovation, Systems and Technologies" series, Springer , 2020
Bengali Informative Chatbot (BIC) is an effective technique that helps a user to trace relevant information by Bengali Natural Language Processing (BNLP). In this research paper, we introduced an algorithmic Bengali Informative Chatbot (BIC) based on information that is significant mathematically and statistically. This work appeared in Proceedings of International Conference on Computer, Communication, Chemical, Materials and Electronic Engineering , IEEE, Bangladesh, 2019.
For the time being, Cricket is an indisputably one of the most interesting game in the world, especially in the territory of South Asian. As human beings are prone to error, sometimes errors have happened to an umpire and about a constant time is taken by the third umpire for an exact decision of a review. The two different domains artificial intelligence and computer vision have become pop in cricket analysis and decision making. This research introduced the successful uses of CNN, Deep CNN, and Inception V3 for detecting third umpire decisions and automated scoring system in Cricket. This work appeared in Proceedings of International Conference on Computer, Communication, Chemical, Materials and Electronic Engineering , IEEE, Rajshahi, Bangladesh, 2019.
This Scientific Research paper is a synopsis of an automated system “Doly: Bengali Chatbot” which gives a reply to a user query on behalf of a human for the education system in the Bengali language. This is an AI-based Chatbot system, mainly based on machine learning algorithms and Bengali Natural Language Processing (BNLP). The machine gets embedded with this knowledge to identify the desired sentences and making a decision within itself, as a response to answer questions. This work appeared in Proceedings of International Conference on Advances in Science, Engineering and Robotics Technology (ICASERT-2019) , IEEE, Dhaka, Bangladesh, 2019.
Md. Kowsher
Natural language processing (NLP) finds huge applications in automation. Lemmatization is an essential preprocessing technique for simplification of a word to its origin-word in NLP. However, there is a scarcity of effective algorithm in Bangla NLP. This leads us to develop an effective Bangla language lemmatization tool. Usually, lemmatization is done by dynamic programming based Levenshtein distance and data structure-based trie (keyword tree) algorithms. In this paper, we have slightly modified the trie algorithm based on prefixes in Bangla language. After that we have developed a mapping algorithm named as “Dictionary Based Search by Removing Affix (DBSRA).” Finally, we have done experimentation for Bangla language lemmatization and our developed DBSRA confirms best performance in compare to other algorithms. This work appeared in International Journal of Intelligent Systems and Applications(IJISA), 2019.
The Question Answering System (QAS) is one of the significant Machine Learning processes that assist a user to find out the relevant information by Natural Language Processing (NLP). In this research paper, we have described the design and implementation of three Bangla intelligence question answering systems (BIQAS) based on mathematical and statistical procedures using question answering data. These procedures are cosine similarity, Jaccard similarity, and Naive Bayes algorithm. The cosine similarity has interacted with dimension reduction technique SVD on user questions and questions answering so that space and time complexity can be reduced. The methodology of this research is separated into three parts: data collection, pre-processing data, and the establishment of a relationship between users’ questions and information. The system’s primary source of knowledge is a collection of the informative data of Noakhali Science and Technology University. We have gotten 93.22% accurate answer by using cosine similarity, 82.67% by Jaccard similarity and 89.32% by Naïve Bayes algorithm. This work appeared in ICCIT, IEEE, 2019
Information Retrieval System is an effective process that helps a user to trace related information by Natural Language Processing (NLP). In this research paper, we present an algorithmic Information Retrieval System(BIRS) based on information and the system is significant mathematically and statistically. This paper is demonstrated by two algorithms for finding out the lemmatization of Bengali words such as Trie and Dictionary Based Search by Removing Affix (DBSRA) as well as compared with Edit Distance for the exact lemmatization. We present the Bengali Anaphora resolution system using the Hobbs’ algorithm to get the correct expression of information. As the actions of questions answering algorithms, the TF-IDF and Cosine Similarity are developed to find out the accurate answer from the documents. In this study, we introduce a Bengali Language Toolkit (BLTK) and Bengali Language Expression (BRE) that make the easiest implication of our task. We have also developed Bengali root word’s corpus, synonym word’s corpus, stop word’s corpus and gathered 672 articles form the popular Bengali newspapers ‘The Daily Prothom Alo’ is our inserted information. For testing this system, we have created 19335 questions from the introduced information and got 97.22% accurate answer. This work appeared in International Journal on Natural Language Computing(IJNLC), 2019.
Type 2 Diabetes is a fast-growing, chronic metabolic disorder due to imbalanced insulin activity. As lots of people are suffering from it, access to proper treatment is necessary to control the problem. Most patients are unaware of health complexity, symptoms and risk factors before diabetes. The motion of this research is a comparative study of seven machine learning classifiers and an artificial neural network method to prognosticate the detection and treatment of diabetes with a high accuracy, in order to identify and treat diabetes patients at an early age. Our training and test dataset is an accumulation of 9483 diabetes patients’ information. The training dataset is large enough to negate overfitting and provide for highly accurate test performance. We use performance measures such as accuracy and precision to find out the best algorithm deep ANN which outperforms with 95.14% accuracy among all other tested machine learning classifiers. We hope our high performing model can be used by hospitals to predict diabetes and drive research into more accurate prediction models. This work appeared in International Conference on Computer and Information Technology, IEEE, 2019.
In Numerical analysis, interpolation is a manner of calculating the unknown values of a function for any conferred value of argument within the limit of the arguments. It provides basically a concept of estimating unknown data with the aid of relating acquainted data. The main goal of this research is to constitute a central difference interpolation method which is derived from the combination of Gauss’s third formula, Gauss’s Backward formula and Gauss’s forward formula. We have also demonstrated the graphical presentations as well as comparison through all the existing interpolation formulas with our propound method of central difference interpolation. By the comparison and graphical presentation, the new method gives the best result with the lowest error from another existing interpolation formula. This work appeared in Journal of Mathematics a nd Statistical Science, 2019.
The main purpose of this research is to find out the best method through iterative methods for solving the nonlinear equation. In this study, the four iterative methods are examined and emphasized to solve the nonlinear equations. From this method explained, the rate of convergence is demonstrated among the 1st degree based iterative methods. After that, the graphical development is established here with the help of the four iterative methods and these results are tested with various functions. An example of the algebraic equation is taken to exhibit the comparison of the approximate error among the methods. Moreover, two examples of the algebraic and transcendental equation are applied to verify the best method, as well as the level of errors, are shown graphically. This work appeared in Journal of Advanced Research in Applied Mathematics and Statistics.
Fraud detection which is a discussible phenomenon to many bounds together with financial sectors, banking, insurance as well as diverse forms of industries. Nowadays fraud endeavors are being amplified with rampant pace especially via the development of technology, so building fraud discovery more significant than ever before. For this reason, we researched and build a fraud detection system using mathematical, statistical and machine learning. This work appeared in Proceedings of International Mathematics Conference, BMS, Bangladesh, 2018.
The main goal of this research is to give the complete conception about numerical integration including Newton-Cotes formulas and aimed at comparing the rate of performance or the rate of accuracy of Trapezoidal, Simpson’s 1/3, and Simpson’s 3/8. To verify the accuracy, we compare each rules demonstrating the smallest error values among them. The software package MATLAB R2013a is applied to determine the best method, as well as the results, are compared. It includes graphical comparisons mentioning these methods graphically. After all, it is then emphasized that the among methods considered, Simpson’s 1/3 is more effective and accurate when the condition of the subdivision is only even for solving a definite integral. This work appeared in Journal of Advanced Research in Applied Mathematics and Statistics, 2019.
The Bengali Informative Intelligence Bot (BIIB) is an effective Machine Learning (ML) technique that helps a user to trace relevant information by Bengali Natural Language Processing (BNLP). In this book, we introduce two mathematical and statistical procedures for BIIB based on information of Noakhali Science and Technology University (NSTU) that is significant mathematically and statistically. In the preprocessing part, this book is demonstrated by two algorithms for finding out the lemmatization of Bengali words such as Trie and Dictionary Based Search by Removing Affix (DBSRA) as well as compared with Edit Distance for the exact lemmatization. We present the Bengali Anaphora Resolution system using the Hobbs‘ algorithm to get the correct expression of consequence questions. In order to reduce the time complexity of searching questions and reply from inserted information, we have used Non-negative Matrix Factorization (NMF) as the topic modeling technique, and the Singular Value Decomposition (SVD) as to reduce the dimension of questions. TF-IDF (Term Frequency-Inverse Document Frequency) has been used to convert character and/or string terms into numerical values, and to find their sentiments. For the action of chatbot in replying questions, we have applied the TFIDF, cosine similarity and Jaccard similarity to find out the accurate answer from the documents. In this study, we introduce a Bengali Language Toolkit (BLTK) and Bengali Language Expression (BRE) that make the easiest implementation of our task. We have also developed Bengali root word‘s corpus, synonym word‘s corpus, stop word‘s corpus, and collected 74 topic related questions and answers from the information of NSTU which are actually our inserted informative questions. For verifying our proposed systems, we have created 2852 questions from the introduced topics. We have got 96.22% accurate answer by using cosine similarity and 84.64% by Jaccard similarity in our proposed BIIB.
Noakhali Science and Technology University
The goal of this research paper is to develop a mathematical and statistical technique that will add an analyzing system for competitive regression in machine learning or data science. In this study, we establish a mathematical solution based on Logistic Growth model for the prediction of regression and competition analysis via competitive data. Generally, when some competitors are going to compete for the same resource after that a competition is being arisen. Then the profit or loss of an agent hardly is being dependent on the policies of other agents or representatives. In this model, an agent can explore the win-loss, evaluate profits, and enforce its policies and anti-agent policies. For this study, we have collected the 10 years renowned SIM Company’s subscribed data of Bangladesh because a high competition is always being expanded among SIM Companies. We describe the graphical solutions of this model (competitive learning) and comparison with other remaining regression methods. We also elaborate a python module for the easiest using of this model.
Springer: Data Science and Engineering.
International Journal of Intelligent Systems and Applications(IJISA)
Machine learning method gives a learning technique that can be applied to extract information from data. Lots of researches are being conducted that involves machine learning techniques for medical diagnosis, prediction and treatment. The goal of this study is to perform several machine learning actions for finding the appropriate mode of birth (cesarean or normal) to minimize maternal mortality rate. To generate a computer-aided decision for selecting between the most common way of baby birth, C-section and vaginal birth, we have used supervised machine learning to train our classification model. A dataset consists of the information of 13,527 delivery patients has been collected from Tarail Upazilla Health complex, Bangladesh. We have implemented nine machine learning classifier algorithms over the whole datasets and compared the performances of all those proposed techniques. The computer recommended mode of baby delivery suggested by the most convincing method named "impact learning," showed an accuracy of 0.89089172 with the F1 value of 0.877871741.
Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering, Springer, 2020.
International Journal of Linguistics & Communication
Journal of Computer Speech and Language, Elsevier.