Sparck Jones, K. (1979) (1979a) Experiments in relevance weighting of search terms, Information Processing and Management, 15, 1979, 133-144. Home/Search Document Not in Database Summary Related Articles Check |
This paper is cited in the following contexts:
TREC2 Document Retrieval Experiments using PIRCS - Kwok Grunfeld (1994) (Correct)
....means performing learning from the known Disk3 (not Disk 1 2) relevant documents before retrieval. In other words, we assume the answer documents are known for training and represent the best that probabilistic theory can provide using our system. This is however not the true upperbound [Spar79] for retrieval from Disk3, because the vocabulary and usage statistics are still those of Disk 1 2. The vocabulary is retained for comparison with routing results. pircs7 retrieval achieves average precision of 0.350, which improves over pircs5 (training from Disk 1 2) by about 10 in average ....
Sparck Jones, K (1979). Experiments in relevance weighting of search terms. Info. Proc. Mgmnt. 15:133-144.
Adaptive Retrieval Agents: Internalizing Local Context and.. - Menczer, Belew (1999) (22 citations) (Correct)
....to improve performance at a local level. Local adaptation is therefore hardwired into the system. Our use of relevance feedback also follows in the mainstream of text retrieval research, perhaps closest to the way in which supervised learning is used to estimate word probability distributions [40, 46]. Again, the main difference is that InfoSpiders use cues provided by user assessments in conjunction with local context; an agent will not waste its limited resources paying attention to a word that never appears in the current search area, even if the user likes that word a lot. Feature ....
K Sparck Jones. Experiments in relevance weighting of search terms. Information Processing and Management, 15:133--144, 1979.
A Probabilistic Model of Information Retrieval.. - Jones, Walker, Robertson (1998) (29 citations) Self-citation (Jones) (Correct)
....information, and useful to have reasonable grounds for believing that estimates based on rather little information, if this is of the right sort, may still be adequate, and hence that even where the incidence data is limited RW s can improve performance. In the earlier experiments reported in Sparck Jones (1979a) the amount of relevance information was systematically increased, reassuringly showing that performance correspondingly improved but also that relatively little information could still be of some value. Sparck Jones (1979b) also reported experiments comparing the use of only a few, randomly ....
....RW s can improve performance. In the earlier experiments reported in Sparck Jones (1979a) the amount of relevance information was systematically increased, reassuringly showing that performance correspondingly improved but also that relatively little information could still be of some value. Sparck Jones (1979b) also reported experiments comparing the use of only a few, randomly chosen relevant documents for prediction as against the full set. These tests also suggested that even a few relevant documents could be helpful, but all of these early tests used older and probably flattering methods of ....
[Article contains additional citation context not shown here]
Sparck Jones, K. (1979) (1979a) Experiments in relevance weighting of search terms, Information Processing and Management, 15, 1979, 133-144.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC