Syed Akhter Hossain

Big data analytics in education: Prospects and challenges

Big data, a widely used buzzword that describes a massive volume of both structured and unstructured data primarily originated from the social media and Web 4.0, has been consistently evolved over the last couple of years through innovation of algorithms and applications for the processing of data. Big data is characterized from the context of volume, velocity and veracity and encompasses different challenges including analysis of data, capture and data curation, efficient storage, transfer and visualization of data along with the privacy and security of data. The analytics defines the process of examining data sets containing a variety of data types to uncover hidden patterns and enable meaningful interpretation for representations and thereafter the applications. This paper will describe the nascent field of big data analytics in education with discussion on prospects and challenges way forward. The paper also intends to focus on research and development issues for educationist and practitioners of big data analytics.

Sentiment Prediction Based on Lexical Analysis Using Deep Learning

This era of computing made everything computerized. So people started to interact in the virtual life more than real life. From shopping to talking, everything is controlled by computer. Because of that it has become more and more important to analyze and extract sentiment. Sentiment can be extracted from different type of data format like audio, text, image. In this paper, we proposed some methods to analyze text data in the paragraph level. Those methods can be implemented using bag of words, and priority was given to the lexical-based analysis.

Ontology-Based Information Retrieval System for University: Methods and Reasoning

The purpose of this paper is to search for a general framework for ontology development. This paper also implemented ontology on university domain, proposed a general framework for ontology searching and explained searching mechanism through university ontology. Also, it presents different ways of reasoning the ontology. In general, ontology classifies the variables in need for some computations and creates interrelationships between them. It is also an essential part of the semantic web. The introduction of semantic web poses the demands for creating ontology in many domains. This paper emphasized mostly on conceptualizing the university as a whole and was developed using standard tools protege 4.3. This paper tries to fill the gap between existing works by including all the concepts and their related data and object properties. The reasoning of our created ontology was done through Fact++ and Hermit 1.3.8 reasoner.

Sentiment Analysis of Tweet Data: The Study of Sentimental State of Human from Tweet Text

Sentiment analysis is a hot topic today. The purpose of this research is finding out sentimental state of a person or a group of people using data mining. The target of this research is building a user friendly interface for general people, so that they will be able to see the analysis report very easily. This analysis process contains both supervised and unsupervised learning, which is a hybrid process. Analysis is done based on keywords, which is defined by the user. User is able to set the number of tweets he/she wants to analyze. We used web-based library for the system. The system is tested and found satisfactory result.

Sentence Similarity Estimation for Text Summarization Using Deep Learning

One of the key challenges of natural language processing (NLP) is to identify the meaning of any text. Text summarization is one of the most challenging applications in the field of NLP where appropriate analysis is needed of given input text. Identifying the degree of relationship among input sentences will help to reduce the inclusion of insignificant sentences in summarized text. Result of summarized text always may not identify by optimal functions, rather a better summarized result could be found by measuring sentence similarities. The current sentence similarity measuring methods only find out the similarity between words and sentences. These methods state only syntactic information of every sentence. There are two major problems to identify similarities between sentences. These problems were never addressed by previous strategies provided the ultimate meaning of the sentence and added the word order, approximately. In this paper, the main objective was tried to measure sentence similarities, which will help to summarize text of any language, but we considered English and Bengali here. Our proposed methods were extensively tested by using several English and Bengali texts, collected from several online news portals, blogs, etc. In all cases, the proposed sentence similarity measures mentioned here was proven effective and satisfactory.

Sentence-Based Topic Modeling Using Lexical Analysis

Data is not meaningful unless its information could be extracted. In every second in this world, we are generating millions of data over the Internet in different form. Most of them are in text format. Usually, data is written based on any topic, or sometimes on few topics. Following this, identifying topic of any text data is very important. Topic identification may help text summarization tools, text classification tool, etc. Machine learning applications may need less training on their data, only if once the topic of text is identified. Therefore, the demand of topic modeling is higher than ever right now. Data scientists are working day and night to make it more effective and accurate using different methods. Topic modeling focuses on the keywords that can express or identify the topic discussed in the document. Topic modeling can save a lot of time by releasing its user from page-to-page manual reviewing. In this paper, a model has been proposed to find out topic of a document. This model works based on the relations between most frequent words and their relation with sentences in the document. This model can be used to increase the accuracy of the topic modeling.

Network architecture and security issues in campus networks

Network architecture with its security is a growing concern in the present time. A campus network faces challenges to address core issues of security which are governed by network architecture. This paper is mainly targeted towards campus networks which deliver required security. This is essential because, it prevents the institution from suffering any significant attacks associated with network. A university network has a number of uses such as teaching, learning, research, management, e-library, and connections with the external uses. Therefore, network architecture and its security are vital issues for any university. In this work, a network infrastructure is proposed on the basis of the practical and experimental requirements. The proposed network infrastructure is realizable with adaptable infrastructure.