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.

BEmoD: Development of Bengali Emotion Dataset for Classifying Expressions of Emotion in Texts

Abstract: Recently, emotion detection in language has increased attention to NLP researchers due to the massive availability of people’s expressions, opinions, and emotions through comments on the Web 2.0 platforms. It is a very challenging task to develop an automatic sentiment analysis system in Bengali due to the scarcity of resources and the unavailability of standard corpora. Therefore, the development of a standard dataset is a prerequisite to analyze emotional expressions in Bengali texts. This paper presents an emotional dataset (hereafter called ‘BEmoD’) for analysis of emotion in Bengali texts and describes its development process, including data crawling, pre-processing, labeling, and verification. BEmoD contains 5200 texts, which are labeled into six basic emotional categories such as anger, fear, surprise, sadness, joy, and disgust, respectively. Dataset evaluation with a Cohen’s \(\kappa \) score of 0.920 shows the agreement among annotators. The evaluation analysis also shows the distribution of emotion words that follow Zipf’s law.

Design of high efficient and stable ultra-thin CdTe solar cells with ZnTe as a potential BSF

Abstract: The polycrystalline ultra-thin cadmium telluride (CdTe) is familiar as the potential solar cell material for its higher efficiency, cost-effective, cell stability and clean generation of solar electricity. In this study, a numerical analysis has been performed utilizing AMPS (Analysis of Microelectronic and Photonic Structures) simulator to examine the cell performances (V oc , J sc , FF and conversion efficiency) of ultra-thin CdTe solar cell. During the research, reduction of CdTe layer was done in the proposed cell and found that 1µm absorber layer is enough for acceptable range for cell conversion efficiency. The possibility of this ultra-thin CdTe absorber layer was examined, as one with 100 nm ZnTe back surface field (BSF) layer to minimize the recombination losses at the back contact and to reduce the barrier height in the valence band of the proposed cell. Higher conversion efficiency of 22.53% (J sc = 24.28 mA/cm2, FF = 0.875, V oc = 1.06 V) has been achieved with only 0.8 µm of CdTe absorber layer along with 100 nm ZnTe BSF where as conversion efficiency is 18.68% (J sc = 21.47 mA/cm2, FF = 0.85, V oc = 1.02 V) without BSF layer. Moreover, the proposed CdTe solar cell showed better stability as the normalized efficiency of the proposed cell linearly decreased with the increasing operating temperature at the gradient of −0.16%/°C.

An artificial neural network based detection and classification of melanoma skin cancer using hybrid texture features

Abstract: Melanoma being the most unpredictable and life-threating cancer, has been on the rise in recent times. In most of the cases being fatal, if treated early, the fatality rate might be lowered severely. Hands-on Melanoma detection at primary stages with the unassisted eye is error-prone and requires vast knowledge and experience. Number of expert dermatologists being inadequate, a computerized and automated approach is needed to accurately detect Melanoma. The following study tries to achieve this feat by developing a neural network that can effectively detect and classify Melanoma. The process begins with preprocessing of dermoscopic images to remove hairs with the Maximum Gradient Intensity algorithm and also enhancement of the images is done. Segmentation based on Otsu Thresholding algorithm is applied to separate skin lesions from the images. Multiple features like ABCD, GLCM, and LBP are then calculated from the segmented images which will be used to train a neural network. The network was successful to attain an accuracy of 97.7% on the combined dataset of ISIC archive the PH2 dermoscopic image database. The proposed method was found to be more accurate than existing methods and encorporates much more feature information from the images.

An overview of Concentrated Solar Power (CSP) technologies and its opportunities in Bangladesh

Abstract: Concentrated Solar Power (CSP) technology is one of the most promising candidate for mitigating the future energy crisis. The extracted power from CSP technology is very clean, reliable and environmental friendly. An overview of CSP technologies like Parabolic Trough, Solar Tower, Parabolic Dish, Linear Fresnel technology have been described in this paper. Comparison among these technologies has been illustrated in this work. For extenuating the present power crisis in Bangladesh CSP technology has great opportunities, since the average Direct Normal Irradiance (DNI) in Bangladesh is 4–6.5 KWh/m2 which is suitable for all types of CSP technology. Suitable locations for different CSP plants in Bangladesh are also proposed on the basis of efficiency, required area and amount of DNI received.