Mohammad Shamsul Arefin

Skyline Query for Selecting Spatial Objects by Utilizing Surrounding Objects

With the increase of data volume, advanced query operators such as skyline queries are necessary in order to help users to handle the huge amount of available data by selecting a set of promising data objects. In this paper, we present a method for selecting spatial objects, such as houses, based on the objects of the surrounding facilities such as restaurants, supermarkets, and stations. In our approach, at first a user specifies a list of surrounding facilities within a specified distance in his preferred location. Our system then computes a set of spatial objects in the preferred location considering the objects of the surrounding facilities by utilizing the idea of skyline queries. We perform different experiments to show the effectiveness of our approach. Experimental evaluation shows that our approach is well applicable for efficient decision making.

Impact analysis of facebook in family bonding

Abstract Nowadays, Facebook is a very popular social communication media. People utilize Facebook to express their thoughts, ideas, poems, and sorrows on Facebook. In the age of information superhighway, majority of the teenagers are not sharing their difficulties, problems, inconsistency, inability and failure with their parents in Bangladesh. However, they share with their friends on Facebook. Subsequently, their friends are making comments, providing shelters and affections to them. Due to lack of education and experiences on technology, guardians in Bangladesh are not aware about the communications and addictions on social Medias. Therefore, there are generating gaps between the guardians and their children. In this paper, a survey-based and Apriori algorithm analyzes the behaviors of teenagers’ by collecting information from their Facebook pages. Parents and teachers opinions are also considered about the activities of students on home and institutes. Here, age limits of targeted teen agers are between 16 and 18. From this analysis, vulnerable relationship between parents and their teenage children have been noticed. The pivotal problem was that teens are spending more time on Facebook and parents want them to the table during study time and school time.

Agent-based anonymous skyline set computation in cloud databases

Given a set of objects, a skyline query finds the objects that are not dominated by another object. A skyline query helps us to filter unnecessary information efficiently and gives us clues for various decision making tasks. On the other hand, we usually have to hide individual records values even though there is no ID information in the table in privacy aware environments. In such situation, we cannot use conventional skyline queries. To handle the privacy problem, we considered a skyline query for sets of objects in a database. In which we do not disclose individual records values. Let s be the number of objects in each set and n be the number of objects in the database. There are

BAENPD: A Bilingual Plagiarism Detector

In this paper, for the first time, we present a bilingual plagiarism detection system, BAENPD that can detect plagiarism from electronic Bangla and English documents.It uses two different methods for detecting plagiarism. Thefirst method is based on the analysis of individual contents ofthe documents, whereas second technique performs several statistical analysis of the documents. The system has been evaluated by real documents. We have found that our system can efficiently detect plagiarism between English and Bangla documents as well as from the documents of same language.

Bangla vowel sign recognition by extracting the fuzzy features

This paper presents a fuzzy based Bangla vowel sign recognition system. Vowel and consonant are the main criteria to express any language. Exceptionally, Bangla has some vowel sign which are used to express Bangla word. Therefore, to recognize any Bangla word, at first it is necessary to recognize which vowel sign is used to express that word. Due to its low computational requirement and very high accuracy, fuzzy logic system is probably the most efficient method available for online vowel sign recognition. The most important task for this implementation is the building of the rule base using fuzzy logic that would describe the recognized vowel sign, which is more complicated as different people write the same vowel sign in completely different ways. In this paper, we propose a system that uses automatically generated fuzzy membership value that describes the handwritten Bangla vowel sign written by different individuals.

Multilingual Content Management in Web Environment

Internet has become primary medium for information access and commerce in todays globalized world and the demographics of the Internet are rapidly turning to be multilingual. Therefore, multilingual content management can be treated as a vital issue for the availability of information in the native languages of the Internet users. To provide the contents to the users in their native languages most of the current approaches keep information for each language and present users information from the chosen language. Developing such systems is both space and time consuming and is not an efficient one. With standard HTML pages, this process involves creating and maintaining multiple versions of each page for each supported language as well as having the language content embedded into the HTML. In this paper, a multilingual content management technique that is based on natural language translation mechanism has been introduced, which can reduce the time and space of the existing systems. In the propose approach, the information is stored in a single language and the contents are translated into users chosen language. The system has been evaluated by both syntactic data and real data obtained from a real life schema. We have compared the performance of our system with existing systems. Our system outperforms the existing systems in terms of both space and time.

An empirical framework for translating of Phrasal verbs of English sentence into Bangla

Generally phrasal verbs comprise a verb followed by a preposition that is commonly occurring feature in English. Each of the phrasal verbs acquires absolutely different meanings in different contexts. Having highly context dependent meanings, phrasal verbs may be disambiguated only by devising a technique involving utilization of semantic information pertaining to the context. This paper presented a framework for translating the phrasal verbs in English sentence into its equivalent Bangla. Semantics is necessary in the field of natural language processing to detect the meaning of the individual words in the sentence. This paper also describes different types of semantic features that perform semantic based disambiguation of phrasal verbs. A translation engine works to find out the exact Bangla meaning of the phrasal verb from different English sentences. We have tested our model for different type of English sentences containing phrasal verbs that are taken from the English newspaper and English textbook. We got successful results for most of the test cases.

BQAS: A bilingual question answering system

In this paper, we present a bilingual question answering system, BQAS that can generate questions and provide their answers from electronic Bangla and English documents. The system first performs the analysis of the contents of the documents to generate questions based on user provided query. It then generates answers of the questions. The system has been evaluated with real documents. We have found that our system can generate questions and related answers up to a satisfactory level from the documents of two different languages as well as from the documents of the same language.

Recommending Hotels by Social Conditions of Locations

In the field of information technology, a recommendation system is a computer program that provides valuable information for the users and guides them to take efficient decisions. The recommendation systems play a vital role in reducing time and effort of users to choose their desired products/services. With rapid growth of Internet technologies recommender systems become very popular to the users nowadays. In this paper, we present a system for recommending hotels for the users. Conventional hotel recommendation systems recommend hotels based on non-spatial attributes of hotels such as price and rating and do not utilize their social locations well. In contrast, proposed system considers the co-existence of other facilities such as restaurants and entertainment facilities in the surrounding areas while selecting a hotel for recommendation. We first evaluate the social conditions of each hotel. Then, we consider user provided reviews about hotels where he stayed earlier. Based on the user’s review, we calculate preferences of that user. Finally, we calculate similarity score between the hotels and the user’s preferences and select the top-k hotels. We perform different experiments to show the effectiveness of our approach. Experimental evaluation shows that our approach is well applicable for recommending hotels for the users.

Developing a framework for ontology generalization

In this paper, we have developed a framework for ontology generalization that can easily generalize two or more similar type ontologies into one ontology. The proposed system first checks the similarity among the ontologies. It then merges the similar ontologies into one more general ontology. We have tested our developed ontology generalization approach with several ontologies. We have found that our system can perform the generalization of ontologies with a certain level of accuracy.