Finding the topics of an article written in Spanish 1 Finding the main themes in a Spanish document * Adolfo Guzmán 1 Centro de Investigación en Computación, Instituto Politécnico Nacional, México aguzman@pollux.cenac.ipn.mx SUMMARY. The computer can easily carry out many operations on systematic collections of data when these are numbers: • What is this data about? What are its main topics? • Make a summary. Obtain a summary of May sales of a given store. • Compare. Compare May sales in stores A and B. • Find similarities and discrepancies. How are sales of stores A and B similar? • Find averages. Find the sales in the South of Mexico, in Fall 1997. • Find tendencies. Extrapolate. On the other hand, when data appears in documents in Spanish, organized in sections, paragraphs and sentences, it is not possible for the computer to carry out the above operations. Since much of human knowledge is in texts written in natural language, it is convenient to discover methods to carry out those operations. For that, the computer must understand or comprehend the text. This paper shows how to analyze a document containing natural language sentences, in order to recognize its main topics or themes. 1. INTRODUCTION AND OBJECTIVES. In the Center for Computing Research of the National Polytechnic Institute, the Laboratory of Natural Language and Text Processing works on intelligent text processing, needed to carry out the operations given in the summary. One of these operations, finding the main topics contained in a document written in Spanish, has been solved. 1.1 The importance of analyzing written Spanish. Intelligent text analysis will allow the computer to understand documents written in natural language, for instance, to summarize them, to find tendencies, to compare two documents (with respect to a given theme) and to answer non trivial questions: Having read this text Answer this question Frogs live in water Do frogs get wet? Benito Juárez is buried in San Fernando Cemetery. Where is the left big toe of Benito Juárez buried? * published in Journal Expert Systems with Applications, Vol. 14, No. 1/2, Jan/Feb 1988, pages 139-148 1 This work was carried out by the autor at SoftwarePro International, owner of all rights on CLASITEX.