Browsing by Author "Hossain, Naimul"

Now showing 1 - 3 of 3

A Deep CNN-based Approach for Detecting Major Disease of Potatoes
(Daffodil International University, 2021-09-11) Hossain, Naimul
Within the advance of the world economy there's continuously a tremendous effect of farming. Considering the situation of Bangladesh 14.74 % of add up to GDP came from the farming segment, where the significance of crops to be more particular crops like potatoes is evident. To overcome misfortune of generation and stabilizing the nourishment chain identifying potato maladies and taking basic steps is required. In that case the mechanization of identifying infections plays a vital part particularly picture handling. Recognizing in a conventional way takes intemperate preparing time and needs skill. Minimizing the preparation time and detecting the infections in early stages is the most objective drawing closer to this issue. The starting step was collecting the information and building a well outfitted information for conveying this work. For conveying this investigative work, this particular issue is drawn nearer with a profound learning strategy. We executed Profound CNN engineering for classification which performs extraordinary in case of picture handling beneath the space of computer vision. The most excellent precision we got from our actualized design is 97.89%.
An Attention Based Approach for Sentiment Analysis of Food Review Dataset
(IEEE, 2020) Bhuiyan, Md. Rafiuzzaman; Mahedi, Mahmudul Hasan; Hossain, Naimul; Tumpa, Zerin Nasrin; Hossain, Syed Akhter
Sentiment Analysis is a technique related to text analysis and natural language processing used to detect various types of insights or information from a portion of text. Over the past few years, researchers have done many works regarding this. In Bangladesh, many online services like-e-com become very popular day by day. One of them is online food delivery services. We can order various foods of our choice from online and sometimes people gives reviews based on that food. Those reviews are usually discarded as unstructured data which of them have no work in further. In this piece of research focus primarily on those unstructured data to analyze them in a correct manner to find insight into customers' behavior and their reactions on those online platforms. To do this experiment first we collect data from websites. Later deep learning-based techniques applied here. For baseline structure, we have used both CNN and LSTM models. Then for improving the model accuracy an attention mechanism applied followed by CNN which gives us 98.45% accuracy. We've also evaluated our model performances with some evaluation metrics also. From them, CNN based attention model gives a higher f1-score of 0.93.
Sentiment Analysis of Restaurant Reviews Using Combined CNN-LSTM
(Scopus, 2020) Hossain, Naimul; Bhuiyan, Md. Rafiuzzaman; Tumpa, Zerin Nasrin; Hossain, Syed Akhter
The combination of machine learning approach and natural language processing is applied to analyze the sentiment of text for particular sentences. In this particular area lots of work done in recent times. Restaurant business was always a popular business in Bangladesh. These business is now Leaning towards online delivery services and the overall quality of restaurants are now judged by reviews of customers. One try to understand the quality of a restaurant by the reviews from other customers. These opinions of customers organizing in structured way and to understand perception of customers reviews and reactions is the main motto of our work. Collecting data was the first thing we have done for deploying this piece of work. Then making a dataset which we harvested from websites and tried to deploy with deep learning technique. In this piece of research, a combined CNN-LSTM architecture used in our dataset and got an accuracy of 94.22%. Also used some other performance metrics to evaluate our model.