Chittagong University of Engineering & Technology

About: Chittagong University of Engineering & Technology is a education organization based out in Chittagong, Bangladesh. It is known for research contribution in the topics: Computer science & Renewable energy. The organization has 1200 authors who have published 1444 publications receiving 10418 citations. The organization is also known as: Engineering College, Chittagong & Bangladesh Institute of Technology, Chittagong.

Early Stage Brain Tumor Detection on MRI Image Using a Hybrid Technique

Abstract: One of the robust issues facing the world today is a brain tumor, out of many diseases in the medical sciences. Brain Tumor is a cellular proliferation in the brain that is uncontrolled. Early detection and diagnosis can reduce the mortality rate significantly. In this paper, a tumor detection algorithm has proposed based on support vector machine (SVM) and artificial neural network. The proposed algorithm is divided into gradient intensity smoothing based edge preservation, image enhancement, SVM based segmentation, features extraction, and classification. To measure the performances of the proposed algorithm is tested on the BRATS dataset image. Experimental result has shown that the proposed algorithm achieved 97.7% accuracy, which is better than many existing algorithms.

A framework for building a natural language interface for Bangla

Abstract: Mobile Computing Devices are enabling connection between people and the Internet, largest source of information in the world. However to properly utilize this knowledge in these devices most people need a Natural Language Interface. Siri, Google Now, Cortana are examples of such interfaces. Because Bangla is a low resource language building such interface is very difficult and time consuming. However due to the increasing numbers of smart phone and smart device users in the Bangla speaking regions, application developers are facing the need for such interfaces to provide web services effectively. This paper addresses this issue and gives an empirical framework on how to build a feasible Natural Language Interface in Bangla and similar low resource languages.

Life cycle cost analysis of solar-LED alternatives to fuel-based fishing net indicator

Abstract: All around the world, most of the developing countries practice ancient technique of fishing where fishing net indicator is an important component of night fishing. In Bangladesh, fishermen use kerosene based lamp which is inefficient, costly and also pollutes the environment. This paper proposes a model which will replace the kerosene lamp with LED light feed from solar energy. This solar-LED system has desirable features such as low cost, zero pollution of the environment and reduces the dependency on fossil fuel by saving a large amount of fuel from a single mechanized boat. Life Cycle Cost Analysis (LCCA) has been analyzed to assess the economical feasibility of the proposed system.

A Modified Naïve Bayesian-based Spam Filter using Support Vector Machine

Abstract: The ever-growing problem which is threatening the current mailing system is spam. Spam is nothing but an unsolicited bulk e-mail frequently sent in a financial nature which generates the need for creating an anti-spam filter. Amongst many spam filtering techniques, the most advanced method "Naive Bayesian filtering" using the Support Vector Machine (SVM) have been implemented. Spammers are very careful about the filtering techniques. For that very reason, dynamic filtering is needed and the proposed method meets the demand. The algorithm splits the received email into tokens and uses Bayes' theorem of probability to calculate the probability of spam for each token to determine the total spam probability of the mail. Implementation of SVM instead of corpora is one of the added features of the algorithm. The most challenging feature was to take the words as well as whole sentences as input in the SVM as tokens and feature vectors. The inclusion of sentences in the dataset training has increased the accuracy of detecting spam and ham. Natural Language Tool Kit (NLTK) has been used as a useful language processing tool to tokenize the sentences and also to understand the meaning of the same types of sentences to some extent. As a test mail is being compared by word to word and also sentence to sentence from the training datasets to determine if the mail is spam or not, it will improve the performance of the filter. With some simple modifications, the filter can be used in both server and client end. The efficiency increases gradually with the increased number of email it processes.

A distributed algorithm to select channel and relay in multichannel-multirelay CR-networks

Abstract: Channel selection in multichannel cognitive radio system and relay selection in multi-relay cooperative system are well studied problems in separate branches. In this study, we consider a joint channel and relay selection problem which is very important to realize multi-relay cooperative transmission in multichannel cognitive radio networks. A joint channel and relay selection algorithm is proposed for amplify-and-forward relaying. A distributed implementation of the proposed joint channel and relay selection algorithm is also presented. As a measure of the channel quality, we consider a crosslayer approach where physical layer channel state information and link layer collision probability of the channels are used as selection parameters. In the joint channel and relay selection, the channels for two phases of cooperative transmission and a single-relay that maximizes the end-to-end mutual information is selected. We show that proposed relaying schemes significantly improve the achievable rate of the system.