The Knowledge Based Search In Wikipedia

Published Date: 02 Nov 2017

The phenomenal growth of information on the web over the last couple of decades leads the way of Wikipedia. Wikipedia is a free internet based Encyclopedia consists and manages the information in the form of articles. In order, to improve the performance of article retrieval while searching in Wikipedia and rather depend on the title of articles, proposed algorithm works to retrieve relevant and accurate information from resources. Dependency on title of articles makes the informational resources limited and cannot be more efficient. As per the experience shared on internet through various resources and online survey, sometimes information is not relevant on Wikipedia which leads the way of a new algorithm which can be more relevant and accurate. Addressing challenges is the objective of this paper is, to make searching algorithm technique used by Wikipedia more efficient and not limited up to classical approach. Idea behind the concept of Knowledge based search in Wikipedia, to make search in Wikipedia for effective results and optimum solutions with a Service Oriented Architecture. Knowledge based searching propose a way more suitable from Title based searching of Wikipedia (classical approach). In knowledge based searching concept defines a way for process or match user’s query with in the titles name and the content under articles for efficient results. The aim of this project is to utilize the information resources in proper way for best results. This project will helps to provide an alternate or better solution for searching information in Wikipedia.

Keywords: Knowledge Based Searching, Classical Searching Algorithm, SOA, Informational Architecture, Semantic Data.

Introduction

Wikipedia is a free internet based Encyclopedia, support multiple languages and works in an open environment. Wikipedia is the best way to get the information. Wikipedia is an open source of information which is supported by Wikimedia. Wikimedia is a non-profit foundation, provides the relevant information in the form of articles. It has become the largest and most prominent general reference work on the Internet. Wikipedia manage information in the form of pages and pages consist of articles. Articles contain information about topics and can be concerned with any subject. Wikipedia’s article are written and managed by the volunteers around the world. Wikipedia has over 25millions articles in 286 languages. There are currently 4,219,959 articles in English Wikipedia alone with total 30,034,144 number of wiki pages. Wikipedia is a website that provides an environment for creation and editing of any number of interlinked webpages [1]. The categories and pages that make Wikipedia is a fertile resource for understanding the development of system and topic. Each page in Wikipedia can be interlinked with multiple categories, formed into a loose ontology of topics. Articles can be managed and edited by anyone who has registered access to the site. Any user can modify the category of article or edit the worthy information. Articles are connect or interlinked together to provide related information. Knowledge based searching, the propose concept can be fit as a role of an alternate way in order to search in Wikipedia. Registration on Wikipedia is optional but mandatory for some specific tasks such as uploading of files, create pages on Wikipedia, editing protected pages etc. Wikipedia is currently available in 275 active editions, with over 77,000 active editors worldwide. The articles on Wikipedia are grown rapidly. Following graph shows the number of article on Wikipedia English edition. Continues Growth of the number of articles in the English Wikipedia reached 3,000,000 articles in August 2009.

The Wikipedia search is restricted up to the Title Based Searching. Title based Search Characterize, it match or compare the user’s search query only with Articles title name, if it matched than it will display the results. Addressing the problem in Title based searching algorithm is it skips the content and priority is determined only for Title, sometime outcome is not so appropriate and cannot utilize the proper informational resources. So deployed Knowledge based searching algorithm can be more relevant to search in Wikipedia. A study conducted by researchers in 2008 regarding content and distribution of topics in each area [2].

Pie Chart of Wikipedia contents and distribution of topics by subject as January of 2008[2].

Schematic Definition of Wikipedia

The description of database during design time is called Schema. How the data will be store in database depend on the schema. Each attribute of schema contains the information about the particular properties of dataset. Following diagram is an xml schema representation of a Wikipedia page containing information about the page, article, user etc.

XML Schema representation of Wikipedia

Following are the details about xml schema of Wikipedia pages

Title tag contains the information about title of articles.

ID and username consists information about the id of particular article and users respectively.

Internal links (or wikilinks) in Wikipedia is managed by [[text]]. A wikilink links a Wikipedia page to another within English Wikipedia. Text will be displayed as link on the browser. [[ ]] is defined as anchor tag in schema of Wikipedia.

Interwiki link links a page with another page on Wikimedia project. It is defined as [[: x]]. Specify the targeted link.

External links are managed as [http://www.xyz.org] display a link which pointed towards an external link on another site.

Linking with in the same page from one section to another can be done by [[pagename#section name | displayed text]].

Some links can be generated automatically like: ISBN, RFC and PMID. There are no need to put these items into square brackets.

Literature Review

There has been many significant research on Wikipedia aimed at characterize and tremendous growth of its content, its evolution, and to understand its topic distribution (see [3] [4]). Halavais and Lackoff [3] compared the topic of distribution quantitatively.

Holloway et al. [4] introduced a graph layout algorithm to put similar articles closed together.

Journal Nature [5] in 2005 proposed that open structure of Wikipedia "makes no assurance of correctness" and had a same level of serious errors as Encyclopedia. Although, this journal reported that the structure of articles in Wikipedia was often poor. Kleinz, Torsten [6] describe that Wikipedia's open structure makes it a target for trolls and waster who add incorrect information to articles.

A summary of literature review on Wikipedia by some researchers

Wikipedia allow its user to develop knowledge actively and collaboratively. (Jaksch, Kepp & Woomer-Hacker, 2008)

Use information for problem solving, research, decision making, and continued professional development. (Orr, Appleton & Wallin, 2001)

Wikipedia considered as a tool that facilitate collaborative finding, shaping, and sharing of knowledge. (Reinhold, 2006)

Proposed Framework

The propose method will helps us to utilize the proper informational resources by not define its limits up to search not only within the title name but also in content or information under articles with utilization of the current technology. Wikipedia is an internet based encyclopedia can be accessed from anywhere all over the world on any platform. The Proposed searching algorithm matches the user’s query with the content of articles as well as with title. The Knowledge Base search searches the subject with all Knowledge Base files, and returns the pages that containing all users search words. The search results page presents user with a list of documents that contain users search terms without replication. This search will skip the one word’s query for example ‘A’, ‘v’ etc. For the English articles (‘a’, ’an’, ’the’) it will search only with the title in the Wikipedia article. Knowledge based searching algorithm depend on both title and content. Addressing challenges with this searching is it should be fast enough to work with a huge dataset. Working environment for this algorithm is very important because maintaining and process query repeatedly on very large data and follow the time and accuracy constraint is very critical. The architecture provides service for application to user and enterprises and manages all the informational resources. Architecture explains the way of provide this services in perspective to enterprises and user’s view. Knowledge based searching algorithm, is a very good idea to change the way of searching the information with title and content of articles which provides a relevant results.

SOA perspective with Knowledge Based Algorithm

The Service oriented architecture is a service offer the way of defining and designing the applications as a discrete unit for business and enterprises. Wikipedia works in an edited environment it’s registered user can modify the valuable information. High level functions, data is provides as services, information exchanged in process, conversion of information from one format or semantics to another are the main part of SOA services. SOA services consider the functional aspects. Service contract can broadly specifies interactions between the service user and service provider contains Service

Service interface

Interface documents

Service policies

Quality of services (QoS)

Performance

This entire service life cycle is managed - from designing, to deployment, to enhancements, to maintenance. The service interface defines the service operations related to information that are passed into and out of the operation to process users query. Interface document is deal with the user and provide result respective to users query. Service policies related to make modification in the Wikipedia documents and provide the qualitative services. Architecture must follow the constraints in order to efficient retrieval of information.

Informational Architecture

Informational architecture can be described into three layers: Physical data (sources), domain (service) data, and semantic data. Physical data stored on the disk, have details of how it is stored and described in database schema. Optimized schema is used for the performance feature and requirement of the particular dataset. Domain data is used in the carrying out service. It represents the knowledge of data. Semantic data describe the information that must be shared between services. Various service interface exchange the semantic data.

Conclusion and Future work

This paper has introduced a new type knowledge based searching which basically related to Wikipedia database. Wikipedia provide domain independent knowledge based searching. This paper focuses on the contents of the articles to make search more efficient and effective. It can be applied with various types of contents, document-collections and various queries to optimize the searching. The future work related this algorithm to make it quicker to work with a large dataset and build a powerful database schema to follow the time and accuracy constraints. This searching gives user to a knowledge based query related result in easy, effective and efficient way. The basic conclusion of this searching is to provide a powerful searching way which is related to topics that you may not have thought of….