Abstract:

Nowadays, Internet is becoming an essential tool in our life. We usually access this network of networks several times a day in order to perform different tasks, included searching for information. The WWW is configured as a huge encyclopedia where we can find any type of relevant material in response to a information need expressed by means of a query formulated in a search engine.
Although most of us do not know, there is a consolidated and active area, from the point of view of research and its application to the industry, named Information Retrieval (IR), which is in charge of dealing with the representation, storage, organization and access of information items.
More and more universities around the World are including in their Computer Science and Information Science undergraduate or graduate degrees courses related to Information Retrieval, where basic or advance knowledge about this area is given, respectively.
Most of these subjects are based on a classic teaching/learning process, which is basically centred on lectures and the state-of-the-art books, the practical use of several web search engines, and, more specifically in Computer Science degrees, the use of existing software libraries to build search engines. But there is a lack of computer-based tools that interactively exemplify the whole process of IR for any type of student, independently of her background.
In this paper, we introduce sulaIR, a tool for supporting the teaching and learning of IR. With a very-easy-to-use computer application, on the one hand, the instructor owns a very intuitive instrument to explain all the stages involved in a search; on the other hand, the student is endowed with an interactive and visual way to understand how these stages are performed step by step, so she can understand most of the main IR concepts.
sulaIR was born from an innovation project at the University of Granada, Spain, with the aim of developing a tool to assist the teaching and learning process, giving as a result a open source application, that covers the main IR phases: document pre-process, indexing, retrieval with vector space and Boolean models and relevance feedback, as well as statistics information about the document collection at hand. In each of the mentioned stages, the user is able to visualize the whole process and is able to interact with the tool to see what it is internally happening.
In order to test the potential role of sulaIR as a computer-aided teaching and learning tool, we performed an evaluation with 30 undergraduate students taking an IR course in the University of Holguín, Cuba. The students were slitted in two groups, test and control, respectively. In the first set, the instructor and students used sulaIR intensively during the whole the course, while in the second, they do not, and used traditional methods. After the course, students were evaluated by means of a written exam. The grades from the students that used sulaIR were grater than those belonging to the control group, fact supported by a hypothesis test. Also, a questionnaire was given to the students from the test group in order to know their opinions about the usefulness and usability of this tool. The results offered some conclusive findings: the tool is very useful to support the teaching and learning process and there is no need of a instructor in order to work with the application autonomously.

Keywords:

Information retrieval, computer-aided learning, evaluation.