2017

Browse

Now showing 1 - 20 of 23

Online Book Store
(Department of Computer Science and Engineering, Islamic University of Technology, Board Bazar, Gazipur, Bangladesh, 2017-11-15) Haidari, Arif; Ahadi, Ahmad
IUT Network
(Department of Computer Science and Engineering (CSE), Islamic University of Technology (IUT), Board Bazar, Gazipur-1704, Bangladesh, 2017-11-15) Hassan, Md. Saif; Chowdhury, Nasir Uddin
Increasing In-Home Physical Activity of Obese Children in Urban Area using Kinect
(Department of Computer Science and Engineering (CSE), Islamic University of Technology (IUT), Board Bazar, Gazipur-1704, Bangladesh, 2017-11-15) Chowdhury, Md. Imran; Sagar, Sadi Mahmud
The benefits of physical exercise extend far beyond weight management. Research shows that regular physical activity can help reduce the risk for several diseases and health conditions and improve the overall quality of life. Research has also demonstrated that virtually all individuals can benefit from regular physical activity. Even among frail and very old adults, mobility and functioning can be improved through physical activity [2]. This paper proposes a framework for in-home physical exercise monitoring based on a Kinect platform. The effectiveness of regular physical exercising has been evidenced in general for preventing the deterioration of chronic diseases and premature death [1]. The analysis goes beyond the state-of-the-art solutions by monitoring more joints and offering more advanced reporting capabilities on the movement such as: the position and trajectory of each joint, the working envelope of each body member, the average velocity, and a measure of the user’s fatigue after an exercise sequence. This data can be visualized and compared to a standard (e.g. a healthy user, for rehabilitation purposes) or an ideal performance (e.g. a perfect sport pose for exercising) in order to give the user a measure on his/her own performance and incite his/her motivation to continue the training program. This research is mainly targeted for children in metropolitan area to continue their daily physical activity but such information can be used as well by a therapist or professional sports trainer to evaluate the progress of a patient or of a trainee.
A Study of Motif Discovery Algorithms
(Department of Computer Science and Engineering, Islamic University of Technology, Board Bazar, Gazipur, Bangladesh, 2017-11-15) Aziz, Abdul; Nadim, Md. Abu Taleb
Virtual Dress Selection for Smarter Shopping
(Department of Computer Science and Engineering (CSE), Islamic University of Technology (IUT), Board Bazar, Gazipur-1704, Bangladesh, 2017-11-15) Noman, Abdullah Al; Mahin, Rafi Mohammad
Until recently, retailers have encountered great difficulties while attempting to sell clothing items via the Internet. Although consumers would like to benefit from the savings and convenience of shopping online, they are unable to determine how the clothes will fit. Businesses would similarly like to increase the proportion of their online sales, which would necessitate fewer physical stores and diminish the possibility of losing profits due to local competitors and returned goods. Our goal is to provide a concept for real time system in details that is able to effortlessly try on countless pieces of clothing, without leaving the comfort of their own homes. Moreover, people can also try to wear good looking dress when they wanted to go out from their home to party or other places. People use mirrors everyday to see how they look and choose clothes they will put on for a day before leaving home. Also in clothing stores, many mirrors are located to help customers for making their decision to buy dress fitting well and looking dresses. In this sense, detail concepts for real time dress up system can answer your questions about dress up as well as the size-fitness of dress without physical don and doff time. The needs for the real time virtual dress up system are obvious. Firstly, benefits for customers are to save don and doff time and estimate their body measurements easily for made-to-measure dress. Customers commonly try on many items and spend lots of time to don and doff to purchase dress. It is very inconvenient for them to take dress items they want to try on, go to a dressing room, take off and put on whenever they find attracting dress. Secondly, shop owners can save costs, because they do not need dressing rooms any more. Additionally, wasting clothes tried on by customers will be reduced.
A Cellular Automata Model for Object Monitoring in Mobile Wireless Sensor Network in Hexagonal Grid
(Department of Computer Science and Engineering (CSE), Islamic University of Technology (IUT), Board Bazar, Gazipur-1704, Bangladesh, 2017-11-15) Dipu, Shahidullah Kaiser; Khan, Md. Rahat Hussein
A Cellular Automata Model for Object Monitoring in Mobile Wireless Sensor Network in Hexagonal Grid By Shahidullah Kaiser (134420), Md. Rahat Hussein Khan (134426) Mobile Wireless Sensor Networks (MWSN) is an ad-hoc network that comprises of a large number of sensors. The sensors usually have limited sensing and communication capabilities. We present a CA model that efficiently monitor a moving object in distributed mobile wireless sensor network. CA is a biologically inspired discrete model. Although there are works that have used CA for developing Object Monitoring in Distributed Mobile Wireless Sensor Network, they mostly used square grids which is not a good representation of the actual MWSN model. We focus on simulating the mobile wireless sensor networks for monitoring a moving object considering energyefficiency. In a general case, a number of objects need to be monitored by the sensors and the objects can be static or mobile. Monitoring mobile objects is considerably harder than monitoring static objects. Mobile sensor networks can be useful in monitoring animal behavior. In this case, animals can move randomly within a network and the mobile sensors can also move to monitor their behavior and other aspects.
Wastage-Aware Routing in Energy-Harvesting Software Defined Wireless Sensor Networks
(Department of Computer Science and Engineering (CSE), Islamic University of Technology (IUT), Board Bazar, Gazipur-1704, Bangladesh, 2017-11-15) Kabir, Md. Rayhan; Abyaad, Rafid
Techno-economic drivers are creating the conditions for a radical change of paradigm in the design and operation of future telecommunications infrastructures. In fact, SDN, NFV, Cloud and Edge-Fog Computing are converging together into a single systemic transformation termed “Softwarization” that will find concrete exploitations in network management. Although wireless equipment manufacturers are increasing their involvement in SDN-related activities, to date there is not a clear and comprehensive understanding of what are the opportunities offered by SDN in most common networking scenarios involving wireless infrastructureless communications and how SDN concepts should be adapted to suit the characteristics of wireless and mobile communications. Here we studied different proposed protocol architecture for software defined wireless network, effective ways for energy efficient WSN and found some of the shortcomings of them. We discuss about some of the challenges facing IOT paradigm and major design requirements as well; with the intention to merge SDN and energy reservation approaches for better performance in terms of network lifetime and latency.
Energy Wastage Detection in Smart Buildings
(Department of Computer Science and Engineering (CSE), Islamic University of Technology (IUT), Board Bazar, Gazipur-1704, Bangladesh, 2017-11-15) Sifat, Samiun-Raji; Shad, Nabil
Energy is wasted due to unconsciousness and negligence of the users. Buildings are the major source of energy consumption. So, most of the energy wastage detection techniques are designed for buildings. In our proposed method we use the occupancy based sensors for the detection of the unnecessary consumption. Along with current data, the stored data is also used for the decision making. To attain accuracy and efficiency, the reinforcement learning algorithm is used in data processing. A user interface is there to take the user feedback. This user feedback is used for the learning growth of reinforcement learning algorithm.
Constraint Based Multicast Routing In Internet of Things
(Department of Computer Science and Engineering (CSE), Islamic University of Technology (IUT), Board Bazar, Gazipur-1704, Bangladesh, 2017-11-15) Rahman, Shifatur; Hasan, Sakib
Multicast routing that meets multiple quality of service constraints is important for supporting multimedia communications in the Internet of Things (IoT). Existing multicast routing technologies for IoT mainly focus on ad hoc sensor networking scenarios; thus, are not responsive and robust enough for supporting multimedia applications in an IoT environment. In order to tackle the challenging problem of multicast routing for multimedia communications in IoT, in this book, we analysed two algorithms for the establishing multicast routing tree for multimedia data transmissions. The proposed algorithms leverage an entropy-based process to aggregate all weights into a comprehensive metric, and then uses it to search a multicast tree on the basis of the spanning tree and shortest path tree algorithms. We went through the evolution of the problem from wired networks to wireless netoworks.The book shows the theoretical analysis and extensive simulations for evaluating the proposed algorithms. Both analytical and experimental results demonstrate that one of the proposed algorithms is more efficient than a representative multiconstrained multicast routing algorithm in terms of both speed and accuracy; thus, is able to support multimedia communications in an IoT environment. We believe that the results shown at the end of the algorithm description are able to provide in-depth insight into the multicast routing algorithm design for multimedia communications in IoT and also will speak for themselves of ensuring a better Multicast Routing for Multimedia networks in Internet Of Things based on multiple constraints.
An Enhanced Community Detection Metric for Weighted and Directed Graph-Based Network
(Department of Computer Science and Engineering (CSE), Islamic University of Technology (IUT), Board Bazar, Gazipur-1704, Bangladesh, 2017-11-15) Kabir, Md. Hasibul; Jahan, Md. Anower
Community detection algorithm tries to find the densely connected units in a large network. For this objective different matrices have come into light. Most of them assume all the vertices in a community belong equally to the community. But these matrices identify the communities as whole, these doesn’t give any information at which content the nodes are connected in a community. Moreover these also face resolution limit for larger networks. For resolving this issue, another matrices named permanence has been applied. But this metric is not defined for weighted and directed graph. Our approach will be to implement the permanence on directed and weighted graph.
Image Classification using Deep Convolutional Neural Networks
(Department of Computer Science and Engineering (CSE), Islamic University of Technology (IUT), Board Bazar, Gazipur-1704, Bangladesh, 2017-11-15) Ahmed, Sabbir; Farhad, Md. Saadman
Image classification is the task of taking an input image and outputting a class (a cat, dog, etc.) or a probability of classes that best describes the image. For humans, this task of recognition is one of the first skills we learn from the moment we are born and is one that comes naturally and effortlessly as adults. Without even thinking twice, we’re able to quickly and seamlessly identify the environment we are in as well as the objects that surround us. When we see an image or just when we look at the world around us, most of the time we are able to immediately characterize the scene and give each object a label, all without even consciously noticing. These skills of being able to quickly recognize patterns, generalize from prior knowledge, and adapt to different image environments are ones that we do not share with our fellow machines. Convolutional neural networks, Sounds like a weird combination of biology and math with a little CS sprinkled in, but these networks have been some of the most influential innovations in the field of computer vision. 2012 was the first year that neural nets grew to prominence as Alex Krizhevsky used them to win that year’s ImageNet competition (basically, the annual Olympics of computer vision), dropping the classification error record from 26% to 15%, an astounding improvement at the time. Ever since then, a host of companies have been using deep learning at the core of their services. In our research we experimentd on image classification using different deep learning frameworks. The following sections describes basics of deep learning and how it can be used in case of image classification and convolutional neural networks. We have also discussed the different deep learning frameworks and current applications. Finally we shared our gained results and knowledge. This will help the researchers to get a clear idea about getting knowledge in the field of image classification with deep convolutional neural networks.
Vibrotactile and Visual Feedback for Deaf and Mute
(Department of Computer Science and Engineering (CSE), Islamic University of Technology (IUT), Board Bazar, Gazipur-1704, Bangladesh, 2017-11-15) Chowdhury, Mehrab Zaman; Siam, Masrur Sobhan
According to World Federation of the Deaf (WFD) there are approximately 70 million deaf and mute all around the world. They are the unfortunate ones who are deprived of communication properly. This creates gap between the normal people and the deprived ones. In our research, we will establish communication between both the normal people and the deprived ones. We will establish one to one and two-way communication. Our research consists of mainly two modules; first module is to generate texts from speech which will then be classified into command sets and context sets, second module is providing reply by the deaf and mute to the normal people. The commands and contexts will be mapped to the database and according to that output will be shown. It is an adaptive and multimodal approach.
Online Shop Management System
(Department of Computer Science and Engineering (CSE), Islamic University of Technology (IUT), Board Bazar, Gazipur-1704, Bangladesh, 2017-11-15) Hasib, Abu Saleh; Opu, Iqbal Hossain
Predicting Breast Cancer Survivability Using Data Mining Techniques
(Department of Computer Science and Engineering (CSE), Islamic University of Technology (IUT), Board Bazar, Gazipur-1704, Bangladesh, 2017-11-15) Abid, Mohammad; Youssouf, Njayou
Breast cancer is one of the major causes of death in women when compared to all other cancers. Breast cancer has become the most hazardous types of cancer among women in the world. In this paper we present an analysis of the prediction of survivability rate of breast cancer patients using data mining techniques. The collection of large volumes of medical data has offered an opportunity to develop prediction models for survival by the medical research community. The data used is the SEER Public-Use Data. The preprocessed data set consists of 262,423 records, which have all the available72 fields from the SEER database. After cleaning of the data set, 106,237 records were put under analysis, then we have investigated five data mining techniques: the Naïve Bayes, the back-propagated neural network, logistic regression, support vector machine and the J48 decision tree algorithms. Comparison of the performance of all these different techniques shows that the Logistic regression has a better performance of 93.07% accuracy. Afterwards, three feature reduction methods, Attribute correlation, Information gain and factor analysis for mixed dataset, were used to reduce data dimension. The result of these methods showed a better performance in time for all the above mentioned data mining techniques. It had a fluctuating accuracy in case of other methods but showed and increase to 94.35% accuracy in case of Logistic regression when factor analysis for mixed dataset was used.
Sensor Based Arm Rehabilitation for Post Stroke Patients
(Department of Computer Science and Engineering (CSE), Islamic University of Technology (IUT), Board Bazar, Gazipur-1704, Bangladesh, 2017-11-15) Kabir, Ridwan; Ehsan, Mohaimin
Noise free data obtained from devices used to track human motion can be used to determine the position, orientation and motion of various parts of human body specially the limbs. These data can be used to determine a proper therapeutic intervention for those people, who face difficulties in moving the different limbs. In our research, we present a method to derive data from sensors like IMU (Inertial Measurement Unit) and Flex sensors and map them to determine the position and orientation of human arm in real time. This will help therapists to ensure proper, accurate and efficient therapeutic intervention for arm rehabilitation of post stroke patients through visualization. The visualization includes a human arm model in 3D space whose position and orientation is determined through forward kinematics using the Denavit-Hartenberg Convention and 3D transformation and rotation
Prediction of a Gene Regulatory Network in Cancer Cells
(Department of Computer Science and Engineering (CSE), Islamic University of Technology (IUT), Board Bazar, Gazipur-1704, Bangladesh, 2017-11-15) Anik, Mustadir Mahmood; Farhan, Nabil
The invention of high throughput technology like microarrays has enabled us to better understand how different cellular components interact. Thus created great interest in the field of Gene Regulatory Network(GRN) in particular. The interplay of interactions between DNA, RNA and proteins leads to genetic regulatory networks (GRN) and in turn controls the gene regulation. Directly or indirectly in a cell such molecules either interact in a positive or in repressive manner therefore it is hard to obtain the accurate computational models through which the final state of a cell can be predicted with certain accuracy. A variety of models and methods have been developed to address different aspects of GRN. Using the Time series data and applying it to these models researchers generate meaningful results i.e. how genes interact with one another. However results found are not of much accuracy due to presence of intrinsic noise of the expression measurements. In order to produce more accurate GRNs using one of the many models available, a new technique is proposed here.
Online Job Recruitment System
(Department of Computer Science and Engineering (CSE), Islamic University of Technology (IUT), Board Bazar, Gazipur-1704, Bangladesh, 2017-11-15) Mohamed, Njoya Pefensie; Moussa, Adama; Sidick, Mbohou Aboubakar
This project is aimed at developing a web page and central recruitment Process system for the HR Group for a company. Some features of this system will be creating vacancies, storing application data, and Interview process initiation, Scheduling interviews, storing Interview results for the applicant and finally Hiring of the applicant. Reports may be required to be generated for the use of the HR group. This project ‘Online Job Recruitment System’ is an online website in which jobseekers can register themselves and then attend the interview. Based on the outcome of the interview the jobseekers will be short listed. The methodoly used to develop our system involved system analysis, system design, system development, and system testing. And it been developed using PHP,HTML and javaScript as programming languages, while XAMPP as server and phpMyadmin as the database of the system.
Music Based Mood Detection
(Department of Computer Science and Engineering (CSE), Islamic University of Technology (IUT), Board Bazar, Gazipur-1704, Bangladesh, 2017-11-15) Ishraq, Shahid; Kamal, Zia Uddin
Music mood describes the inherent emotional expression of a music clip. It is helpful in music understanding, music retrieval, and some other music-related applications. In this paper, a hierarchical framework is presented to automate the task of mood detection from acoustic music data, by following some music psychological theories in western cultures. The hierarchical framework has the advantage of emphasizing the most suitable features in different detection tasks. Three feature sets, including intensity, timbre, and rhythm are extracted to represent the characteristics of a music clip. The intensity feature set is represented by the energy in each sub-band, the timbre feature set is composed of the spectral shape features and spectral contrast features, and the rhythm feature set indicates three aspects that are closely related with an individual’s mood response, including rhythm strength, rhythm regularity, and tempo. Furthermore, since mood is usually changeable in an entire piece of classical music, the approach to mood detection is extended to mood tracking for a music piece, by dividing the music into several independent segments, each of which contains a homogeneous emotional expression. Preliminary evaluations indicate that the proposed algorithms produce satisfactory results. On our testing database composed of 800 representative music clips, the average accuracy of mood detection achieves up to 86.3%. We can also on average recall 84.1% of the mood boundaries from nine testing music pieces. A method is proposed for detecting the emotions of song lyrics based on an affective lexicon. The lexicon is composed of words translated from ANEW and words selected by other means. For each lyric sentence, emotion units, each based on an emotion word in the lexicon, are found out, and the influences of modifiers and tenses on emotion units are taken into consideration. The emotion of a sentence is calculated from its emotion units. Tofigure out the prominent emotions of a lyric, a fuzzy clustering method is used to group the lyric’s sentences according to their emotions. The emotion of a cluster is worked out from that of its sentences considering the individual weight of each sentence. Clusters are weighted according to the weights and confidences of their sentences and singing speeds of sentences are considered as the adjustment of the weights of clusters. Finally, the emotion of the cluster with the highest weight is selected from the prominent emotions as the main emotion of the lyric. The performance of our approach is evaluated through an experimentof emotion classification of 400 song lyrics. Therefore, the main idea is to merge the two ideas altogether.
BCI Text Entry Us ing Hierarchical Keyboard With Probabilistically Dynamic Clustering
(Department of Computer Science and Engineering (CSE), Islamic University of Technology (IUT), Board Bazar, Gazipur-1704, Bangladesh, 2017-11-15) Hayet, Ishrak; Haq, Tanveer Fahad
The ability to feel, adapt, reason, remember and communicate makes human a social being. Disabilities limit opportunities and capabilities to socialize. With the recent advancement in brain computer interface (BCI) technology, researchers are exploring if BCI can be augmented with human computer interaction (HCI) to give a new hope of restoring independence to disabled individuals. This motivates us to lay down our research objective, which is as follows. In this study, we propose to work with a hands free text entry application based on the brain signals, for the task of communication, where the user can select a letter or word based on the intentions of left or right hand movement, and left, right, up or down nodding movement. The three major challenges that have been addressed are (i) interacting with only four imagery signals (ii) how a low quality, noisy EEG signal can be competently processed and classified using novel combination of feature set to make the interface work efficiently, and (iii) using a language prediction model to increase characters per minute.
MIH Secondary School
(Department of Computer Science and Engineering (CSE), Islamic University of Technology (IUT), Board Bazar, Gazipur-1704, Bangladesh, 2017-11-15) Alkasim, Hussain; Haruna, Ibrahim; Said, Muhammed