The principle that, for optimal retrieval, documents should be ranked in order of the probability of relevance is discussed. (Author/KP)

It would be very helpful in retrieval experiments if good retrieval performance for a test collection was known, so that performance for particular devices could be fully evaluated. This paper presents one performance yardstick, based on optimally weighted request terms, and illustrates its application to different test collections. (Author)

Farradane's relational indexing appeared applicable to the scientific area of the properties of metals and gave good results with a greater depth of indexing. (Author/PF)

Presents a new concept-based method to analyze the text characteristics of documents at varying relevance levels. Applies the results of the document analysis in an experiment on query expansion in a probabilistic information retrieval system and investigates statistical differences in textual characteristics of highly relevant and less relevant…

Poses questions concerning methods of using specificity information in a weighting scheme. (JB)

The work described was designed to test the feasibility of using automatic or semi-automatic methods to replace or reduce the intellectual effort involved in retrieving information from machine readable stores, particularly those using natural language. (Author)

Methods of testing systems in practice and in theory are critically reviewed. Some new theoretical considerations are advanced. (Author)

Explores the possibility of combining the two principles of ranking according to degree of relevance, and ranking according to probability of relevance, for information retrieval systems, but concludes that, while neither is adequate alone, no single, all-embracing ranking principle can be constructed to replace the two. (Author/VT)

The inverse relationship of recall and precision is not a fundamental law. It must not be used as an excuse for failing to bring about any overall improvement in the performance of information retrieval systems (12 references) (Author/NH)

The present experiments were designed to study the effects of search term weighting based on very limited relevance information, supplied, for example, by one or two relevant documents. The tests simulated iterative searching as in an on-line system and show that even very little relevance information can be extremely valuable. A list of…

A model for an information retrieval system in which recall and precision are inversely related is described, and then the anomaly is resolved. (PF)

A test was carried out to investigate the assumption that there is a significant level of redundancy between documents. The conclusions were that notional overlap exists, but there appear to be very few cases of extreme redundancy, and at the document level, there is little possibility of practical application. (Author)

Some of the factors that must inevitably affect the assessment of printed indexes are discussed, together with the observation of two unexpected phenomena, namely an apparent relationship between precision and searching effort, and relationships between generality and both recall and fallout. (4 references) (Author)

An investigation was made into the relative performance of the four major Chemical Abstracts Service magnetic tape data-bases. The overall retrieval efficiency of profiles searched against data-bases containing titles only, titles-plus-keywords, and titles-plus-digests, was calculated and these results are presented. (5 references) (Author/NH)

Reports the results of experiments using colon classification for the analysis, representation, and retrieval of primary information from the full text of documents. Recall, precision, and search length measures indicate colon classification did not perform significantly better than Boolean or simple word occurrence systems. Thirteen references…