Kelledy, F.; Smeaton, A.F.: Thresholding the postings lists in information retrieval : experiments on TREC data (1995)
0.01
0.009384007 = product of:
0.02815202 = sum of:
0.02815202 = product of:
0.05630404 = sum of:
0.05630404 = weight(_text_:indexing in 5804) [ClassicSimilarity], result of:
0.05630404 = score(doc=5804,freq=2.0), product of:
0.19018644 = queryWeight, product of:
3.8278677 = idf(docFreq=2614, maxDocs=44218)
0.049684696 = queryNorm
0.29604656 = fieldWeight in 5804, product of:
1.4142135 = tf(freq=2.0), with freq of:
2.0 = termFreq=2.0
3.8278677 = idf(docFreq=2614, maxDocs=44218)
0.0546875 = fieldNorm(doc=5804)
0.5 = coord(1/2)
0.33333334 = coord(1/3)
- Abstract
- A variety of methods for speeding up the response time of information retrieval processes have been put forward, one of which is the idea of thresholding. Thresholding relies on the data in information retrieval storage structures being organised to allow cut-off points to be used during processing. These cut-off points or thresholds are designed and ised to reduce the amount of information processed and to maintain the quality or minimise the degradation of response to a user's query. TREC is an annual series of benchmarking exercises to compare indexing and retrieval techniques. Reports experiments with a portion of the TREC data where features are introduced into the retrieval process to improve response time. These features improve response time while maintaining the same level of retrieval effectiveness
