Free Method Report Sample

Information Retrieval

Information Retrieval
The information source that I used was Olympic Org (http://www.olympic.org/); this is the Olympic Movement's official website. Olympic movement organizes conducts and regulates all the international Olympic sporting events. The events involve more than 200 nations and are held every four years in varying and different cities around the world alternatingly. As such, an information retrieval system is an absolute must for the Website considering that an individual could be seeking information from any of the previous Olympic events or even just about a particular athlete.

Olympic.org provides two means of searching the website, and the primary means is an obvious search box at the top-left corner of the site. The other method was a link to search “all results and Olympic medalists” To fully and thoroughly analyze the information retrieval system of the site I made my searches in both the options.
I used two informational tasks in the evaluation of the information retrieval system of the site. These tasks relied on the sets provided. I searched for Usain Bolt, an athlete who rose to prominence two years ago because of his outstanding speed in the competition. The basis of my search was the fact that I know that Usain Bolt exists and that has won numerous medals, therefore, information about him on the official website of the Olympic movement was virtually guaranteed.
My second informational task was a location search. Considering that there are two Olympic events, Summer Olympics, and Winter Olympics; I sought the location of the next summer Olympics. My particular query was, “Where is the next summer Olympics?” My rationale for picking this search query was that the location of any future Olympics must be imperative to the movement, and they should, therefore, ensure that anyone who wants to find out its location obtains it efficiently.

Results

Primary Information Retrieval; The search box
In my first task, the search of Usain Bolt the results was numerous encompassing term result pages. The information, however seemed to be a result appeared to be a simple list of all the documents (web pages) on the website with the keyword Usain Bolt. In addition, the search term was highlighted in yellow. The purpose of the highlighting seemed to provide a reference for the searcher to identify its context on the page using the preceding and following words. The search results were, therefore, mainly news and articles that contained information about Bolts achievements in various events.
The search for the location of the next summer Olympics was less helpful with the results being keyword based. The ranking system of the search engine gave precedence to the page with the greatest number of keywords, from the search. My search term was, “Where is the next summer Olympics” Consequently, the first result was the page that had the words is, where, the and next. The aforementioned first search result was a page with information about, “Best of Vancouver Exhibition” I iterated the search by searching, “next summer Olympics” The iterated search was more efficient. The first page containing the list of results informed me that the next summer Olympics are in Brazil in 2016. The Second result had information regarding the Summer Youth Olympics that are in Singapore.

Second Information Retrieval Option in the Website

The second retrieval system was similar to the advanced search offered by the search engines. It provides an avenue for the user to narrow down his search to the specific information that the user is seeking. The user picks from two options, the option for searching a game, sport or event and the option for searching for a medalist. The options have drop down lists that allow one to choose the athlete’s country and the Olympic game. As previously mentioned, I searched for Usain Bolt; the results displayed methodically in a table format. The table had columns of the game (the location and year), the sport, the event, the results (score), the medal won and photos and videos of the event.

Analysis

The information retrieval system is rudimentary and only functions by searching the keywords; this impairs its ability to deliver relevant results consistently. In my search for the location of the next summer, Olympic competition the information retrieved was useless in consideration of the search term. The ranking system of the result also lacked; as a result that ranked first had nothing to do with the search query. The retrieval system of the website is clearly not semantic and requires careful wording of the search term or query to obtain relevant results.
The retrieval system also failed to recognize that word where was not a part of the word required in the search results but simply a question. The lack of recognizance indicates lack of Natural Language Processing in the search engine. Such limitations make it hard for the users who are unable to change their terms to account for a keyword-only search engine.
The nature of the results ensure that the only results obtained are documents rather than information. Such a retrieval system is flawed because an individual might be seeking specific information rather than the whole document. For example my search for Usain Bolt in the search box just needed his profile, the informational retrieval system on the other hand provided me with numerous web pages containing the keywords, “Usain Bolt” but no information about the content.
The information retrieval system also failed in recognizing the error in the query “Usian Bilt” the results found no matching terms. The failure demonstrates lack of any AI engine behind the informational retrieval system of the website. The lack means that the search engine is inflexible and cannot adapt to error and generate relevant results despite the presence of these mistakes.
Nonetheless, the informational retrieval system does have its positive aspects. The main advantage is that serves the simple search queries well. The documents that contained Usain Bolt results displayed, and if I were just seeking any of them then, my search would be done. In addition, the second information retrieval option allowed the results to be narrowed down efficiently. In summary, the information retrieval system in the Olympic.org might be far from perfect but it may serve the minor demands placed on it.
It is paramount to realize that the information retrieval system has a rudimentary capability of discerning slight spelling errors. This is to mean that the system will recognize the errors that are not ridiculous or completely off. Such a capability will often mitigate the errors that a user might make but still do not suffice.

Recommendations

First, there must be improvements in the keyword-based results; this paper has already pointed out the deficiencies of such a system. The information retrieval system must be capable of utilizing more than a word matching program to a more semantic process that can obtain the information sought. Furthermore, the natural language processing, or at least an element of it must be introduced to the information retrieval system. The purpose and aim of this is to allow the queries with words such as where or when to be treated as guiding questions rather that a part of the results. The retrieval system should also produce results with priority on the information sought instead of the documents; rather the results should be content oriented rather than document or files oriented.
The results also ought to be specialized, while the website has attempted to have specialized search for the medalists and the events it has failed to include other possible searches. The site user should have the option of searching for the multimedia, the images, the profiles and even the games and their respective events. Such a system would ease the process of searching for a particular event or even the multimedia. For example, one can be searching for the video of opening of the 1988 Winter Olympics only. If the information system has a process, for instance, a drop-down list where the user selects that their search only has videos then a search of “opening ceremony 1988” would provide the video result. Presently the process is tedious and hampered, as the user has to visit the documents that are on the results page to discover whether they contain his video. In addition, a non-specific search can then generate aggregated results. My search of Usain Bolt would have, therefore, generated the results that included the Profile of the athlete, his events and finally the news on him. Such result would ensure that I obtained the most relevant information about Usain Bolt, and if I needed more, I could resort to a more specific search engine.

References

Buttcher, Stefan, et al. Information Retrieval: Implementing and Evaluating Search Engines. Massachusetts: MIT Press, 2010. Print.
Chapman, Wendy W, et al. "Overcoming barriers to NLP for clinical text: the role of shared tasks and the need for additional creative solutions." Journal of the American Medical Informatics Association (2011): 540-543. Online.
Gorvardhan, A, G Madhu and Rajinikanth. "Intelligent Semantic Web Search Engines: A Brief Survey." The Semantic Web (2011): 234-239. Print.
Kowalki, Gerald. "Information Retrieval System Functions." Information Retrieval Architecture and Algorithms. Springer (2011): 1-26. Online.
Lewandowski, Dirk. The Retrieval Effectiveness of Search Engines on Navigational Queries. New York: Emerald Group Publishing Limited, 2011. Print.
Liu, Tie-Yan. Learning to rank for information retrieval. New York: Springer, 2011. Print.
Maleki-Dizaji, Saeedeh S, Jawed Siddiqi and Fazilatur Rahman. "Adaptive information retrieval system via modelling user behaviour." Journal of Ambient Intelligence and Humanized Computing (2014): 105-110. Online.
Setchi, Rossitza, Qiao Tang and Ivan Stankov. "Semantic-based information retrieval in support of concept design." Advanced Engineering Informatics (2011): 131-146. Online.
Valacich, Joseph S, Christoph Scheneider and Leonard M Jessup. Information systems today: managing in the digital world. Boston: Pearson, 2014. Print.
Zoeter, Onno, et al. Information retrieval system. USA: Patent 8,037,043. 11 October 2011. Online.