Browsing by concept:
another way of approaching the inaugural lecture
another way of approaching the inaugural lecture
Summary
1. The use of ICTs in a text which analyses the use of ICTs...
Advances in computational linguistics, an emerging discipline which combines linguistics with fields closer to formal sciences (mathematics) or technology (computing), have been spectacular in recent years, especially in terms of its applications: for some time now, applications such as automatic translations systems, voice recognition programmes or automatic text summaries have been part of our reality.
Precisely through the automatic use of texts, it has been possible to create "another way" of approaching the inaugural lecture this year: consequently, for example, the post-edited result of this use shows us what the pairs of related concepts are (such as ICT-school, computers-students, virtual university-traditional university) that occur most frequently in the text; it shows us what the most frequently occurring specialist concepts are; and, finally, based on all this information gathered, we have developed a pre-configured search which enables us to discover what we can find on the Internet about the concepts selected.
Conclusions? Many and none: this alternative approach, albeit not experimental and using only some of the possibilities available to us today, must (far beyond statistical curiosities) allow each reader to draw their own conclusions. According to their interests: more formal (discourse, lexical and statistical analysis) or more in terms of content (about concepts and how to express them).
Consequently, being able to "browse" the inaugural lecture based on these parameters may make us notice the emphasis placed on certain aspects by the author, the use of some terms and not others, and also, now that we have an original text in English with Catalan and Castilian versions, it can make us see how the ideas are expressed differently - not always with the same names - in each version.
In short, of course, based on the understanding that the only "valid" reading (even the only "good" one) is the one that can be made of the first to the last pages of Professor Martin Carnoy's text. In fact, this browsing by concept will be particularly beneficial once the whole text has been read, and read well.
2. Browsing by concept: how?
2.a. How can we browse by related concepts?3. Final note
The index of pairs of concepts (in the right-hand margin of the text) enables us to discover:By clicking on every phrase found, we can access the exact place where it appears in the original text.
- what the concepts are which occur most frequently together in a phrase;
e.g. ICT and school;- and what the phrases are where both concepts appear;
e.g. "The use of ICTs, just like any other subject covered in school, depends on the aptitudes of teachers".
2.b. How can we browse by the most frequent concepts?
If what we want is to discover what we can find on the Internet based on the concepts that appear most frequently in the inaugural lecture, we simply select the interest boxes and activate the search. The result will be a search that takes into account the concepts selected, in Catalan, in Castilian and in English.
Finally, we would like to point out that in the text there are, naturally, many more concepts apart from those which we have offered as most frequent or as related pairs. We should bear in mind that the frequencies with which the words-concepts appear or appear together do not necessarily reflect their importance in the text.
This does not, we feel, invalidate the fact that this different, always complementary, approach to the text provides us with an alternative reading of the text by taking into account aspects of the discourse which would be scarcely or not at all visible in conventional reading.
Credits
This work is by Quim Moré and Antoni Oliver, computational linguists at the UOC Language Service, who have been responsible for the computerised use of texts and the resulting analysis and post-editing. Website application and layout by programmer Néstor Postigo. Text revision by linguist Mercè Vázquez. And, finally, Lluis Rius was responsible for coordinating all areas and for the final edition of the paper.