other,0-1-H01-1001,ak <term> Oral communication </term> is ubiquitous and carries important
tech,16-3-H01-1001,ak information piece would be found in a <term> large database </term> . Traditional <term> information retrieval
tech,1-4-H01-1001,ak large database </term> . Traditional <term> information retrieval techniques </term> use a <term> histogram </term> of <term>
model,6-4-H01-1001,ak information retrieval techniques </term> use a <term> histogram </term> of <term> keywords </term> as the <term>
other,8-4-H01-1001,ak </term> use a <term> histogram </term> of <term> keywords </term> as the <term> document representation
other,11-4-H01-1001,ak </term> of <term> keywords </term> as the <term> document representation </term> but <term> oral communication </term>
other,14-4-H01-1001,ak <term> document representation </term> but <term> oral communication </term> may offer additional <term> indices
model,19-4-H01-1001,ak communication </term> may offer additional <term> indices </term> such as the time and place of the
model,2-5-H01-1001,ak and the attendance . An alternative <term> index </term> could be the activity such as discussing
other,18-7-H01-1001,ak Similar to activities one can define <term> subsets </term> of larger <term> database </term> and
tech,21-7-H01-1001,ak define <term> subsets </term> of larger <term> database </term> and detect those automatically which
model,3-8-H01-1001,ak database of TV shows . Emotions and other <term> indices </term> such as the dominance distribution
tech,6-9-H01-1001,ak directly . Despite the small size of the <term> databases </term> used some results about the effectiveness
model,15-9-H01-1001,ak results about the effectiveness of these <term> indices </term> can be obtained . To support engaging
skrij podrobnosti