A Review Of Journal Article On Search Engine Operations (Part 3)

You are welcome to my blog today. This is the third and final part of the post: A Review Of Journal Article On Search Engine Operations as you can read the previous part here, Enjoy reading and have a wonderful day.. Kumari, Gupta & Dixit (2014) in their research paper discussed the page rank algorithm.The paper highlights the strengths, weaknesses. They explained that the Page Rank algorithm uses link structure to determine the importance of web page. This algorithm is based on random surfer model. According to them, the random surfer model assumes that a user randomly keeps on clicking the links on a page and if she/he get bored of a page then switches to another page randomly. Thus, a user under this model shows no bias towards any page or link. Page Rank (PR) is the probability of a page being visited by such user under this model. Page Rank algorithm assumes that if a page has a link to another page then it votes for that page. Therefore, each in-link to a page raises its importance. Page Rank is a recursive algorithm in which the Page Rank of a page depends upon the Page Rank of the pages linking to it. Thus, not only the number of in-links of a page influences its ranking but also the page ranks of the pages linking to it. A page confers importance to the pages it references to by evenly distributing its Page Rank value among all its  out-links.  Kumari, Gupta & Dixit (2014) gave the working formula that the Page Rank algorithm uses to rank a page. The formula is stated as follows.
PR(P) = 1 – d + d
·        Where N0...Nn are the pages that point to page P.
·        O(Ni) is defined as the number of links going out of page P.
·        The parameter d is a Damping factor which can be set between 0 and 1.
They further explained that damping factor, d is the probability of user’s following the direct links and 1- d denotes the rank distribution from non– directly linked web pages. It is usually set to 0.85.
Kumari, Gupta & Dixit (2014) further outlined the strengths and weaknesses of the Page Rank algorithm. The strengths included:
·        Less Time consuming:- As Page Rank is a query independent algorithm i.e. it computes the rank score ahead, so it takes very less time .
·         Feasibility:-This algorithm is more feasible as it computes rank score at indexing time not at query time.
·        Importance: - It returns important pages as Rank is calculated on the basis of the popularity of a page.
·         Less susceptibility to localized links: - For calculating rank value of a page, it consider the entire web graph, rather than a small subset, it is less susceptible to localized link spam.
The weaknesses as highlighted by them include:
·        The main disadvantage is that it favors older pages, because a new page, even a very good one, will not have many links unless it is part of an existing web site.
·        Relevancy of the resultant pages to the user query is very less as it does not consider the content of web page.
·        Dangling link: This occurs when a page contains a link such that the hypertext points to a page with no outgoing links. Such a link is known as Dangling Link.
·        Rank Sinks: The Rank sinks problem occurs when in a network pages get in infinite link cycles.
·        Dead Ends: Dead Ends are simply pages with no outgoing links.
·        Spider Traps: Another problem in Page Rank is Spider Traps. A group of pages is a spider trap if there are no links from within the group to outside the group.
·        Circular References: If you have circle references in your website, then it will reduce your front page’s Page Rank.
Wenpu Xing and Ali Ghorbani projected a Weighted Page Rank (WPR) algorithm which is a modification to the Page Rank algorithm. This algorithm assigns a larger rank values to the more important pages rather than dividing the rank value of a page evenly among its outgoing linked pages. Each outgoing link gets a value proportional to its importance. The importance is assigned in terms of weight values to the incoming and outgoing links.  The strengths of the weighted page rank include:
·        Quality: Quality of the pages returned by this algorithm is high as compared to Page Rank algorithm.
·        Efficiency: It is more efficient than Page Rank because rank value of a page is divided among its out-link pages according to importance of that page.One noticeable weakness about this algorithm is that as this algorithm considers only link structure not the content of the page, it returns less relevant results to the user query. I have these posts series have given you a better understanding of search engine operations. Have a nice time and Thank you for reading.