How to search

The corpus is composed of four sections:

  1. Conversations
  2. Search by intervention
  3. Search by intonational group
  4. Search by Keyword

1. Conversations

This searching allows a complete reading of the transcripts. Likewise, if the researcher needs to do some further work, conversations can be exported directly to Word, Excel or XML.

2. Search interventions

Search for intervention is designed to look up words present in the interventions. Therefore, it is mandatory to enter a word in the "Text into the intervention." In this case, it is possible to filter speaker features. This field has enabled a possible language search with MYSQL.

For example, if you enter the word "tener" in the search field, the results will include interventions that not only appears "tener" but "tenerte, tenerlo, mantener, sostener...". To find only "tener" can be searched in the following sequence:

% tener %

For more information: http://dev.mysql.com/doc/refman/5.0/es/pattern-matching.html

In any case, if the aim is to make a more specific search, it is recommended to search by keyword.

3. Search by intonational group

Search by intonational group has the same requirements as the search for intervention, but it has been added other factors such as the presence of adjacent tones pauses or intonation group. As in the previous case, allows a code sequence search with MySQL.

4. Search by word

The keyword search allows searching crossing intonational group and intervention features in which the word appears. Likewise, there is a search word lexical distance located at a maximum distance of 2 words forwards and backwards. In the case of lexical distances, it should be noted that the search should be performed considering that the information in these fields is arranged as follows:

WORD,LEMA,TAG

For example, to look "eres muy listo," can be written "eres" in the "word" and the lexical distance field (+1) could be sought considering that typing "muy," will be the word, but you can also look for "muy," to search the lema or "RG" (EAGLE Freeling label for positive adverb). Thus, for "eres", there is a field "lexical distance (+1)" in which the information would be included as follows "muy,muy,RG". Theres is no space before the commas.

In other words, the search by word, lema or category is splitted into three fields to search by just one word, but if it is required to find more words (lexical form, lema or tag) located at a distance of 2 onwards or backward, this data is joined on these fields.

English