File Information
File: 05-lr/acl_arc_1_sum/cleansed_text/xml_by_section/concl/97/p97-1066_concl.xml
Size: 2,301 bytes
Last Modified: 2025-10-06 13:57:51
<?xml version="1.0" standalone="yes"?> <Paper uid="P97-1066"> <Title>Knowledge Acquisition from Texts : Using an Automatic Clustering Method Based on Noun-Modifier Relationship</Title> <Section position="8" start_page="505" end_page="505" type="concl"> <SectionTitle> 5 Discussion </SectionTitle> <Paragraph position="0"> * Evaluation of the quality of the clustering procedure * in the majority of the works using clustering methods, the evaluation of the quality of the method used is based on recall and precision parameters. In our case, it is not possible to have an a priori reference classification.</Paragraph> <Paragraph position="1"> The reference classification is highly domainand task-dependent. The only criterion that we have at the present time is a qualitative one : that is the usefulness of the results of the clustering methods for a KE building a conceptual model. We asked the KE to evaluate the quality of the clusters, by scoring each of them, assuming that there are three types of clusters : 1. Non relevant clusters.</Paragraph> <Paragraph position="2"> 2. Relevant clusters that cannot be labelled.</Paragraph> <Paragraph position="3"> 3. Relevant clusters that can be labelled.</Paragraph> <Paragraph position="4"> Then an overall clustering score is computed.</Paragraph> <Paragraph position="5"> This elementary qualitative scoring allowed the KE to say that the clustering obtained with the second data set is better than the one obtained with the first.</Paragraph> <Paragraph position="6"> LEXICLASS is a generic clustering module, it only needs nominal (or verbal) compounds described by dependancy relationships. It may use the results of any morpho-syntactic analyzer which provides dependancy relations (e.g. verb-object relationship).</Paragraph> <Paragraph position="7"> The interactive conceptual analysis : in the present article, we only described the first step of the KA process (the &quot;conceptual fields&quot; construction). Actually, this process continues in an interactive manner : the system uses the conceptual fields defined by the KE to compute new conceptual structures; these are accepted or rejected by the KE and the exploration of both the terminological network and the documentation continues.</Paragraph> </Section> class="xml-element"></Paper>