Competency E
design, query, and evaluate information retrieval systems
Without an efficient retrieval system, information would languish unused. A library without a catalog, an index, or any type of searching tool would be a mere collection of books and stacks of documents. Librarians and information professionals are in charge of collecting information, describing it, indexing it, and storing it in a way that can allow for its satisfactory on-demand retrieval. However, in order to achieve this we must first be able to effectively design, query, and evaluate information retrieval systems. As demonstrated in the assignments and description of my comprehension of this competency which follow, in the MLIS program I have been fortunate to learn and develop the skills needed to translate this competency into real world applications.
I have been exposed to information retrieval systems as a user since I 'discovered' the Internet when I was 17 but my first professional exposure really took place in the LIBR 202 (Information Retrieval) class. In this class, I had a chance to explore the principles of information retrieval and their application to information systems and services. I was also able to participate in the design, creation, testing, and evaluation of small databases which are the basis of information retrieval systems.
In the LIBR 202 class I learned that there are several steps in the design of an information system. Before even starting to build a database it is important to have an understanding of the collection that will be indexed as its organization will dictate how information should be accessed. Next the user and the way in which he or she will use the system must also be defined. Other considerations such as deciding what points of access the system will offer, articulating the metadata that will be used to describe the items that are being represented, and creating the vocabulary control are the next logical steps.
I appreciated a group assignment in indexing and vocabulary design where I helped develop an information retrieval system for peer-reviewed articles in the field of Library and Information Science. This project shows my understanding of the structure of information systems and my competency designing, performing queries for testing purposes, and analyzing search results. For the assignment we were required to use the Inmagic's DB/TextWorks tool, a database and text retrieval software that enables one to build networked and stand-alone text databases to manage diverse types of information including documents, images, and multimedia. Our group selected, read, and organized fifteen journal articles to be included in the database. Next, we defined the purpose of the database (to offer an index of journal articles in the LIS field) and our user group ( MLIS students or anyone looking for information in the field who were assumed to already be familiar with the concepts and vocabulary of information storage and retrieval). From there we defined the textbase structure (fields, points of access), created pre-coordinate and post-coordinate controlled vocabularies, created the database in DB/TextWorks, developed a user guide, and put together the layout of the ideal interface. My role in this group assignment consisted of reading the articles and developing keywords to describe them accurately. I also helped enter the raw data into the DB/TextWorks database and designed the look of the interface. This exercise proved to me the complexity of retrieval system design. I realized that information retrieval systems have to be carefully designed in order to be truly efficient with one of the most important components being the design of their controlled vocabularies. Finally, this class also showed me that the best retrieval systems must be designed taking into account different points of view, group discussions, user/usability studies and testing, and feedback. This is what allows for the refinement of the system which ultimately determines whether it is an effective one or not.
A measurement of database efficacy is its success in a real world application. Is a database fairly easy to search? Is it designed efficiently so that results can be easily retrieved? It is important for librarians and information professionals to know and understand how to best present their queries in order to obtain good search results. In LIBR 244 (Online Searching) I studied the best techniques for conducting queries and search strategies, learned to identify the most appropriate database to search depending on the type of information needed, and gained knowledge in querying and evaluating databases. The following search assignments show my competency in querying and evaluating search results in DIALOG, Nexis, and Internet search engines. In each exercise I explain the thinking and evaluation process behind my search strategies as framed by the structure of the engine being used. In a fun final assignment I compared all three databases with regards to ease of use, quality of information retrieved, and number of results obtained.
The strategies and experience gained in this course have allowed me to immediately transfer this skill set to real life applications. For example, in my present job I designed and maintain on an ongoing basis a database of investors that organizes all of their financial investment information with the company. It is a fairly complex dynamic SQL and Microsoft Access based database that includes a litany of information that has, at the same time, close relationships with other information sets. For example, when a query is executed in this database, a user has the flexibility to include as little or as many fields as he or she desires. A sample of the fields can include investor name, investor address, investor committed capital, investor banking information, and investment capital remaining to be called in by our company. The results, which change according to the user's query, can then be exported to a more user friendly format such as a spreadsheet or comma separated values. I constantly utilize this database to execute queries and obtain the data that allows me to recall information for phoen inquiries or process quarterly reports, capital calls, and end of year statements electronically without having to manually comb through our archives. Designing this program was much easier once I had taken these classes as I had a deeper understanding of what a database is, what it is supposed to do, and how it can better support a user's searching experience.
Another professional application for which the skill set acquired in this course has helped me tremendously is web site design. Though I had taken HTML and Dreamweaver classes outside of the program that had taught me how to design aesthetically pleasing Internet pages, it was not until I took these courses that I understood how to properly optimize them for indexing by search engines. For example, the last couple of web sites (www.blueprintventures.com and www.powerfitcamp.com) I have designed contain carefully thought out metadata fields within their HTML coding. These are the fields that describe, very concisely, what a web page is about. Most search engines use this information, as well as keyword searching and links to URLs, to improve recall and the precision of users queries. This is proven by executing a search for both business names without using Boolean operators (to make the search more imprecise) and noticing that both businesses are listed within the first page of search results in Google, one of the most powerful search engines available.
Through my classroom work and professional experiences I have learned that librarians and information professionals need to know how to effectively design, query and evaluate information retrieval systems so that we can better meet the information needs of our users. It is obvious to me now that information retrieval systems, whether a card catalog at a library or an Internet search engine, are the most essential tools we have for accessing information. Even if I never have to design a large information retrieval system, my job as a librarian or information professional will in all likelihood require my evaluating and recommending search systems both to users and management alike, something that I feel confident doing after taking these courses. As well, I now understand how information retrieval systems work, which has helped me use them more effectively to execute searches and ultimately help other users.
Evidentiary
LIBR 202 - Information Retrieval - Information Retrieval System Design
LIBR 244 - Online Searching - Dialog
LIBR 244 - Online Searching - Nexis
LIBR 244 - Online Searching - Dialog and the Internet
LIBR 244 - Online Searching - Dialog, Nexis,and Internet Comparison