Context Based Personalized Search Engine For Online Learning

Published on December 2016 | Categories: Documents | Downloads: 41 | Comments: 0 | Views: 319
of 6
Download PDF   Embed   Report

The WWW serves as a huge widely distributed, global information service centre for acquiring on line knowledge. But there are lots of challenges in finding up the effective, relevant information. During the past few years, it has become apparent that there is an urgent need for tools that guide Web users in their information finding and navigational activities. One of the failings of traditional information retrieval models is attributed to the isolation of queries from the context in which they occur. Mounting a system that can understand the user’s seeking needs beforehand and then presents a list of ranked links to the web pages most probably having the relevant information according to user’s context to lift up information retrieval process that meets the user’s requirement in a effective manner. With this objective, Context Based Personalized Search Engine for online Learning supports the hypothesis that the relevance of search results can be improved through the inclusion of contextual information. This contextual information can be obtained by an automated analysis of the local information space surrounding a given candidate page with the local space defined by the navigational structure inherent the Web.

Comments

Content

(IJCSIS) International Journal of Computer Science and Information Security, Vol. 8, No. 7, October 2010

Context Based Personalized Search Engine For Online Learning
Dr. Ritu Soni , Prof. & Head, DCSA, [email protected] Mrs. Preeti Bakshi, Lect. Compuret Science, [email protected]

GNG College, Santpura, Haryana, INDIA-13
Abstract: The WWW serves as a huge widely distributed, global information service centre for acquiring on line knowledge. But there are lots of challenges in finding up the effective, relevant information. During the past few years, it has become apparent that there is an urgent need for tools that guide Web users in their information finding and navigational activities. One of the failings of traditional information retrieval models is attributed to the isolation of queries from the context in which they occur. Mounting a system that can understand the user’s seeking needs beforehand and then presents a list of ranked links to the web pages most probably having the relevant information according to user’s context to lift up information retrieval process that meets the user’s requirement in a effective manner. With this objective, Context Based Personalized Search Engine for online Learning supports the hypothesis that the relevance of search results can be improved through the inclusion of contextual information. This contextual information can be obtained by an automated analysis of the local information space surrounding a given candidate page with the local space defined by the navigational structure inherent the Web. Keywords: WWW, web personalization, Context Management, Conceptual Architecture, CFL, IR 1. Introduction: In order to satisfy human quest for knowledge, self paced, On-line learning is gaining its importance. The World Wide Web [1] serves as a huge widely distributed, global information service centre for news, advertisements, customer information, financial management, education, government, e- commerce and many other information services. The Web also contains a rich and dynamic collection of hyperlink information and Web page access and usage information. . Given that there is this vast and ever growing amount of information, how does the average user quickly find what she is looking for a task in which the present day search engines don't seem to help much! This is because of high complexity of Web page collection, lack of unifying structure with respect to indexing by category or by title, author, cover page, table of contents etc, highly dynamic nature, broad diversity of user communities that uses only 1% of relevant information, non-effective & inefficient simple keyword based search engine for Web resource discovery and the requirement for adding value to e-services on the Web, necessity towards the creation of loyal visitors customers for a Web site also adds to the disadvantage of Web usage.[2][3] However, based on the above observations, the web also poses great challenges for effective resource and knowledge discovery. How can the portion of the Web that is truly relevant to the user’s interests be determined? How can we find high quality Web pages on a specific topic? These challenges have promoted research into effective discovery and efficient use of resources on the internet. One possible approach is to personalize [4] the web space. “Web personalization is simply defined as a task of making Web based information system adaptive to the needs and interest of users”. In other words a system is created which responds to user queries by potentially aggregating information from several sources in a manner which is dependent on who the user is.[5] The techniques of mining & the user’s access patterns have been applied in a wide range of application including Web search engine and Web personalization. Search engines [6] are generally affected by problems ambiguity and results ordered by website popularity rather than user interests. As most of the present search engines treat the search requests in isolation, they create an index of words within documents and return a rank list of documents in respond to user queries. However only few of the results, returned by the search engines may be valuable. The document is valuable to the user or not depends upon the context of the query asked. It is very important to have dedicated search engine for specific study. Such an application will save user’s time from searching from unstructured web. “The proposed model generates links to the user’s query by understanding the context in which query is initiated and then personalizing user’s request according to the user’s interest area so that the information needs of the user is satisfied” is the idea behind the Proposed Model.

239

1

http://sites.google.com/site/ijcsis/ ISSN 1947-5500

(IJCSIS) International Journal of Computer Science and Information Security, Vol. 8, No. 7, October 2010

The idea of Personalizing the Information Retrieval in User context is achieved by considering and combing the two different but related concept: a) Personalization and b) Context Management.[7] Personalization is a Process of presenting the right information to the right user at the right moment. [8] It is achieved through creating user profile, which gives the description of user interest, preference etc. Information about the user can be collected in two ways: i) Explicitly, for example asking for the feedback such preferences or ratings; and ii) Implicitly, for example observing user behavior such as the time spent reading online document. The presented model follows the first way i.e. it creates the user profiles by explicitly gathering the information about the user preferences. Based upon this personalization information search for requested query is contextualize and system presents better results to the user. It is common knowledge that several forms of context exits in the area. 2. Objective of Study During the past few years, it has become apparent that there is an urgent need for tools that guide Web users in their information finding and navigational activities. One of the failings of traditional information retrieval models is attributed to the isolation of queries from the context in which they occur. Developing a system that can understand the user’s seeking needs beforehand and then presents a list of ranked links to the web pages most probably having the relevant information according to user’s context to lift up information retrieval process that meets the user’s requirement in a effective manner. The above stated is the main objective of our work. Our work supports the hypothesis that the relevance of search results can be improved through the inclusion of contextual information. This contextual information can be obtained by an automated analysis of the local information space surrounding a given candidate page with the local space defined by the navigational structure inherent the Web. To obtain the above stated goal our work exploits the contextual information and smoothly integrates it into the personalization of information retrieval. The idea of contextual personalization [9] proposed and developed here respond to the fact that human preferences are complex, multiple, heterogeneous, changing even contradictory and should be understood in context with the user’s goal and task

at hand. Context is difficult notion to grasp and capture in a software system. Several researchers have tried to categorize context-aware applications and features including the Contextual sensing, Contextual adaptation, Contextual resource discovery and Contextual augmentation. [10] We have focused our efforts on the major topic of information search and retrieval system by restricting it in context of user’s usage access patterns while browsing the Web. The approach behind this work is alleviating information overload offered by the Web on the user and making the Web a friendlier environment for its individual user and hence creating trustworthy relationship between the Web-site and the visitor-customer. 3. Proposed Model The model presents here is concerned with exploiting semantic, ontology based contextual information, specifically aimed towards its use in Personalization for information retrieval. The goal of the presented model is to endow personalized system with the capability to filter and focus their knowledge about user preferences on the semantic context of ongoing user activities. This paper aims at creating a system, that is (semi) automatically tailored, for the content delivered to the user, from a web site. The proposed system constitutes both the contents, as well as the users' interaction. By applying mining methodologies on the Web data one can discover or understand the user’s interest area or his/her habits of browsing the web. The study involves the process of getting user’s interest and needs, analyzing and organizing this valuable information in the context in accordance to user’s requirement, and then providing the user, the result in the form of links that connects to the Web pages containing the needed detail. 3.1 Components for proposed system Conceptual Architecture of the proposed system has components for front–end tools for semantic based web services & administration tools; required functionality tools and data storages & ontology with proper security. The core components for proposed system provide the required functionality for semantic web enabled web services [11] for on line learning. These components are as follows: Administrator: It controls, co-ordinates & administer all other components. Message Transformation: This component supports the semantic transformations at different levels of data, logics & rules, protocol engines etc.

240

2

http://sites.google.com/site/ijcsis/ ISSN 1947-5500

(IJCSIS) International Journal of Computer Science and Information Security, Vol. 8, No. 7, October 2010

• Web Search management: This component enables the search of a particular web page in a particular user profiles in administration process. Auditing / Tracking: It provides trust to the different users through administrator. • User Manager: It has complete user profile. Security: This component provides security for authorization & authentication of user. Web Services & Goal Manager: This component allows managers to standardize & simplify data access while hiding the complexity of processes. Protocol Engine: This allows message passing • over network. Deployment: It makes Web services available for the user in different platforms. Adapter: They provide the adaptability to the future technologies so as to reuse the objects. Transport: It provides network and protocol connectivity.

Fig 1 Conceptual Architecture

only the Administrator who can make the changes in the Site functioning and in Data Base. The Proposed Model is based on the explicit approach of personalizing search in user Context. Here user is required to mark and submit his/her interests in the form of preferred links; it is not done automatically by the system. Thus the process of extracting user interest is user dependent. Only the authorized user can have an access to the system. However any kind of user can register him/her as the authorized user just by filling up a registration form. There will be no charges or fee for the membership. Boundary Conditions: The proposed model is design in two views. One is for the Users and Second is for the Administrator of the Site. The user provides an access only to the User’s View. It is only the Administrator who can make the changes in the Site functioning and in Data Base. • A number of techniques has been evolved and developed in the area of Context Management as explained in the chapter 4. The Proposed Model is based on the Explicit approach of Personalizing search in user Context i.e. it requires user to enter his/her interest to the system. In the Proposed model user is required to mark and submit his/her interests in the form of preferred links, it is not done automatically by the system. Thus the process of extracting user interest is user dependent. • Only the authorized can have an access to the system. It requires that users must have an account with the system to avail its services. However any kind of user can register him/her as the authorized user just by filling up a registration form. There will be no charges or fee for the membership. The hypothesis depicted in the given study is applied only to the limited volume of the data, to show the effect that how context management enhance the results of information retrieval process and provides benefits to the user. Module chart of the Proposed System

Fig 2 CFL Semantic web architect

The proposed model can also be explained through different layers of CFL Semantic web architect as described above. 3.2 Model Design: The proposed model is structured as a search engine that makes the use of consideration described above. It records the user preferences in the form of links navigated while browsing the site in the logs and create user profiles for their regular users. Whenever that particular visits the site again these logs are accessed to personalize the search in user’s context. The proposed model is structured in two views one is User View and second is Administrator View. The user has an access only to User View. It provides trust on the web site with the help of authorization and authentication rules. Users can retrieve information according to their requirements, from the data stored. This is possible by user manager component. The Administrator View is only accessible by the administrator for the management of the model designed. Functional and technical tools are developed using asp.net. • 3.4 Model Requirements: The proposed model is design in two views ie Users view and Administrator of the Site. The user provides an access only to the User’s View. It is

241

3

http://sites.google.com/site/ijcsis/ ISSN 1947-5500

(IJCSIS) International Journal of Computer Science and Information Security, Vol. 8, No. 7, October 2010

4. Implementation: Taking the consideration of personalizing search in user’s context as explained in the previous section of the study the proposed model is implemented as a Web search engine. The Implemented model has two views. One is User view and the Second is Administrator View. The View available for the visitors is the User View. The Administrator View having different domain name, is kept hidden from the users and is accessible to administrator only. The administrator can use this view for different analysis purposes or to upgrade the system services or can use it for different decision making purposes. 4.1 The User View: The processes under user view are explained below. This is the start page of site.

by navigating the links. The user is supposed to manage its profile by Add/Update account. This will help in personalizing the search for the next session. Some more options are available to view the managed links and to contact to the Administrator. User can close his account by sign out. Each screen has the link to go to the home page. User can also place its search experience through the suggestion link. The user can navigate the site through the following links: Home: To come to the home page of the Site. Registration: To register yourself as a authenticated user of the Site. Login: To get the access of the site service. Topics: To see the topics and the related links about which the system has store the data. Suggestions: To get the feedback from the user. Contact: Help to contact with the administrator. Details of some of the user view’s components are as follows: • User’s Welcome Window: This window appears when the existing user performs the login process. Actually it is the home page of the user’s window.

Brief functioning of user view: The option Topics provides only the general view of the topics and underlying links. It is only the general view , search will not initiate from here. The system will open only to the registered user through the login process. After this search proceeds from the User’s Welcome Window. Form here user can select the required topic and can get the required information

Process Description: After the user has successfully finished the Login process, the system will show the User’s welcome window to the user. From here the users can select the topic of their interests and can find the information for which they are looking for. This window also provides some more function to the user to manage his/her account or interests. The functions are listed below: Home: It will show the same window presented above. View Managed Links: This will show the list of topics and links navigated by the users in the previous session.

242

4

http://sites.google.com/site/ijcsis/ ISSN 1947-5500

(IJCSIS) International Journal of Computer Science and Information Security, Vol. 8, No. 7, October 2010

Change Password: This will lets user to change the password. • Link Navigation: The window shows the list of links related to a particular topic.

Process Description: From here the user can select any number of links and navigate them to find the information he/she seeking for. To manage the links according to the user’s context the system requires that the particular user mark the interested links and then add or update this information to his/her profile by clicking on the Add/Update Account button on the lower left corner of the window. The system creates or maintains the user profile for the purpose of personalizing the search in the user’s context to gain the benefit in the Information Retrieval (IR) Process. When the same user visits the same topic area again, then these profiles help the system in providing the user a list of rank links according to his/her interests. • View Managed Links: This window presents list of topics and links marked by the user as interests in the previous session of the visit to the site. This process shows a list of rank links marked by the user as interested links in previous session of his/her visit. The user can start his navigation process here. This provides an advantage to the user, as it saves lots of time of user spending in finding the required or interested links he/she wants to check out.

Process description: 4.2 The Administrator View: This view is accessed by the administrator only. He can make changes to the system to meet the users’ requirement. The processes under administrator view are listed below. This form is the administrator start page. This page will display after the successful Login process. Administrator View also presents various functions to the Administrator that is listed below.

Home: This will show the Administrator’s home page. Feedback: This will show the suggestions given by the users. Users: This will show the list of the users of the site. Topic: This will add a topic to the topic list. Links: This will add link to the specified topic. Deletion: This will delete a topic or a link under a topic. Management: This will give the list of user and the links navigated and marked by them as their interest areas. Password: This will change the password of the Administrator. 5. Conclusion Context is an increasingly common notion in Information Retrieval. Several authors in the IR field have explored approaches that are similar to one presented. The distinctive aspect in our

243

5

http://sites.google.com/site/ijcsis/ ISSN 1947-5500

(IJCSIS) International Journal of Computer Science and Information Security, Vol. 8, No. 7, October 2010

approach is the use of semantic concepts, rather than plain terms (i.e. keywords), for the representation of these contextual meanings, and the exploitation of explicit ontology-based information attached to the concepts, available in a knowledge base. This extra, formal information allows determining the set of concepts than can be properly attributed to the context, in a more accurate and reliable way (by analyzing explicit semantic relations). This paper proposes an approach to improving the precision of search engines, whilst maintaining the level of recall achievable, by utilizing the local information space (as defined by linked pages) in providing a search context. To demonstrate the validity of the approach we have develop a simple model of search engine. This model- even though based on a very simple model of context - is sufficient to provide statistically significant enhancement of the relevance of search results. We are currently carrying out further tests to quantify more precisely the level of improvement that is achievable. 6. Future Scope To improve this approach further, there are a number of directions for further work. The most significant of these is to develop an improved model of the relationship between information context and hyper-textually linked pages in a web environment. This model forms the foundations of the approach being adopted and is required for developing a more rigorously supported set of searching algorithms. 7. References [1] World Wide Web Consortium (W3C) Semantic Web activity homepage. http://w3c.org/sw.rg/wc/robots.html. [2] Berners-Lee T. The Semantic Web and Challenges. W3C Web site Slideshow 2003. URL: http://www.w3.org/2003/Talks/01-swebtbl/Overview.html [3] Berners-Lee T. Standards, Semantics and Survival. SIIA Upgrade 2003:pp 6–10. [4]. Sergio Flesca, Sergio Greco,Ester Zmpano, ”Mining User Preferences, Page Count and Usage to Personalize WebSite Navigation”, August 2005 [5] P.H Mylonasi, D. Vallet, P. Castell, M. Ferna and Y. Avrit Hi Si ,”Personalized information retrieval based on context”, 2008, The Knowledge engineering Review, Vol. 23:1,73-100, Cambridge university Press. [6] Antoniou G. and von Harmelen F. A Semantic Web Primer. The MIT Press,Cambridge, massachusetts, London, England 2004. ISBN 0262-01210-3.

[7] Ajay D Kshemkalyani, George Samaras and Andrew Citron, “Context management and its applications to distributed transactions”, December 1996 [8] Michael Sintek and Stefan Decker. Triple - a query, inference, and transformation language for the semantic Web. In Proceedings of the First International Semantic Web Conference on The Semantic Web, Sardinia, Italy, 9-12 June 2002. Springer-Verlag. [9] Ian Horrocks and Sergio Tessaris. “Querying the semantic Web: A formal approach”, In Proceedings of the 1st International Semantic Web Conference, Sardinia, Italy, 9-12 June 2002. Springer-Verlag. [10] A Stephen J.H. Yang a, Jia Zhang b, Irene Y.L. Chen c, “ JESS-enabled context elicitation system for providing context-aware Web services”, The Journal of system and software 2007. [12] D Fensel, C. Bussler, and A. Maedche, “ Semantic web enabled services” , proceeding of International Semantic Web Conference 2002, LNCS, Springer, pages 1-2,2002 [13] A Ankolenkar, M. Burstein, T. C. Son, J. Hobbs, O.Lassila, D Martin, D McDermott, S. Mcllraith, al et. “DAML-S : Semantic Markup Web Services” Online available at http://www.daml.org/services /.2001 [14] M.Shaw and D Garlan: “Software Architectures” Prentice-Hall, 1996. [15] G. Wiederhold. Mediators, “Architecture of future information systems”. IEEE Computer, 25(3), 1992. [16] D. Fensel and R. Groenboom. A software architecture for knowledge-based systems. “The Knowledge Engineering Review (KER)”, 14(3), 1999.

244

6

http://sites.google.com/site/ijcsis/ ISSN 1947-5500

Sponsor Documents

Or use your account on DocShare.tips

Hide

Forgot your password?

Or register your new account on DocShare.tips

Hide

Lost your password? Please enter your email address. You will receive a link to create a new password.

Back to log-in

Close