scispace - formally typeset
Search or ask a question

Showing papers on "Ranking (information retrieval) published in 1985"




Journal ArticleDOI
05 Jun 1985
TL;DR: The results indicate that while the absolute performance of a search on a particular collection is dependent on the pairwise similarity of the relevant documents, the relative effectiveness of clustered retrieval versus sequential retrieval is independent of this factor.
Abstract: A new means of evaluating the cluster hypothesis is introduced and the results of such an evaluation are presented for four collections. The results of retrieval experiments comparing a sequential search, a cluster-based search, and a search of the clustered collection in which individual documents are scored against the query are also presented. These results indicate that while the absolute performance of a search on a particular collection is dependent on the pairwise similarity of the relevant documents, the relative effectiveness of clustered retrieval versus sequential retrieval is independent of this factor. However, retrieval of entire clusters in response to a query usually results in a poorer performance than retrieval of individual documents from clusters.

183 citations


Journal ArticleDOI
TL;DR: An algorithm for optimal data collection in random fields, the so-called variance reduction analysis, which is an extension of kriging, is presented, which shows a high degree of stability with respect to noisy inputs.
Abstract: This paper presents an algorithm for optimal data collection in random fields, the so-called variance reduction analysis, which is an extension of kriging. The basis of variance reduction analysis is an information response function (i.e., the amount of information gain at an arbitrary point due to a measurement at another site). The ranking of potential sites is conducted using an information ranking function. The optimal number of new points is then identified by an economic gain function. The selected sequence of sites for further sampling shows a high degree of stability with respect to noisy inputs.

130 citations



Journal ArticleDOI
TL;DR: The speed of the performance-index ranking algorithm is estimated for several large network problems, and found to be orders of magnitude higher than that of its direct though more versatile alternative, explicit linearized contingency evaluation.

77 citations


Journal ArticleDOI
TL;DR: Lower bounds of n2/2 and 3n2/4 messages are proved for a worst case execution of any algorithm to solve the ranking and sorting problems, respectively.
Abstract: We study the problems of sorting and ranking n processors that have initial values (not necessarily distinct) in a distributed system. Sorting means that the initial values have to move around in the network and be assigned to the processors according to their distinct identities, while ranking means that the numbers 1, 2,..., n have to be assigned to the processors according to their initial values; ties between initial values can be broken in any chosen way. Assuming a tree network, and assuming that a message can contain an initial value, an identity, or a rank, we present an algorithm for the ranking problem that uses, in the worst case, at most n2/2+O(n) such messages. The algorithm is then extended to perform sorting, using in the worst case at most 3n2/4+O(n) messages. Both algorithms are using a total of O(n) space. The algorithms are extended to general networks. The expected behavior of these algorithms for three classes of trees is discussed. Assuming that the initial values, identities, and ranks can be compared only within themselves, lower bounds of n2/2 and 3n2/4 messages are proved for a worst case execution of any algorithm to solve the ranking and sorting problems, respectively.

56 citations


Proceedings ArticleDOI
05 Jun 1985
TL;DR: The GRIPS retrieval language contains thesaurus, operators to extend the query and truncation — and context-operators for freetext and Boolean searching, and several other viewpoints and evaluation measures were defined and applied in the retrieval experiment.
Abstract: Besides the operators 'and', 'or' and 'not' the GRIPS retrieval language contains thesaurus — operators to extend the query and truncation — and context-operators for freetext and Boolean searching.In a similar way several other viewpoints and evaluation measures were defined and applied in the retrieval experiment. Under the assumption that the evaluation measure is an interval scale averaging is done by calculating the arithmetic mean.As levels of the experimental factor /8/ the following similarity measures were used: inner product measure, cosine measure, overlap measure, coefficient of Jaccard and Euclidean distance. As situative factors /8/ the number of documents retrieved by GRIPS, the number of descriptiors of the queries, generality and topic of documents were used.The retrieval experiment is not yet finished completely but several results have already been obtained. For example in the average (over 81 queries) for the above defined viewpoint the ranking with the inner product measure does not indicate a significant improvement compared with the GRIPS-output. In the case of the Euclidean distance measure it seems that in the average the user has to inspect less nonrelevant documents. This means an improvement compared with the unordered retrieved set from the GRIPS-output.For more details of the so called 'two-level retrieval process' and further experimental results we refer to the long version of this paper.

33 citations


Proceedings Article
29 Oct 1985
TL;DR: An apparatus for applying AC power to a load while maintaining a minimum charge across a capacitor for use as a DC power supply, that may be connected in series with the power supply and load.
Abstract: An apparatus for applying AC power to a load while maintaining a minimum charge across a capacitor for use as a DC power supply, that may be connected in series with the power supply and load, and does not require a parallel connection to the power supply, including power control means responsive to the phase of the power supply signal such that the voltage across said power control means is allowed to rise to a given voltage each cycle, before power is applied to the load, and charging means operable to charge said capacitor to said given voltage.

19 citations


Journal ArticleDOI
TL;DR: In this paper, the backward arcs and the backward length of a paired comparison digraph are defined and the optimal ranking of the vertices of the digraph is defined, which is a ranking whose backward length is minimum among those of all rankings.
Abstract: A paired comparison digraph $D = ( V,A )$ is a weighted digraph in which the sum of the weights of arcs, if any, joining two distinct vertices is exactly one; otherwise, there exist no arcs joining them. A one-to-one mapping $\alpha$ from V onto $\{ 1,2, \cdots , | V | \}$ is called a ranking of D. We define the backward arcs and the backward length of $\alpha$. An optimal ranking of D is a ranking whose backward length is minimum among those of all rankings of D. Our method of ranking the vertices of D is one that makes use of these optimal rankings. For certain classes of paired comparison digraphs, we show that the optimal rankings can be explicitly computed.

17 citations


Journal ArticleDOI
TL;DR: In this article, the problem of ranking contestants in a round-robin tournament has been studied and a general class of models of competition which have been proposed to solve this problem and which lead to the usual method of ranking by number of victories.
Abstract: In a December 1983 article in Management Science, Stephen Goddard studies the problem of ranking contestants in a round-robin tournament He gives objections to the usual method of ranking contestants by number of victories and proposes both an alternative method for ranking contestants and a criterion for determining the “fairest” ranking We argue here that his criterion is faulty and, in fact, falls prey to his own objections We also show that his method, in any event, produces rankings which don't satisfy the criterion Finally we describe a general class of models of competition which have been proposed to solve this problem and which lead to the usual method of ranking by number of victories

Journal ArticleDOI
TL;DR: In this article, a ranking of journals based on their relative quality and impact as perceived by members of the American Risk and Insurance Association is presented, which is based on the AIA's Journal Citation Index.
Abstract: The purpose of this study is to provide a ranking of journals based on their relative quality and impact as perceived by members of the American Risk and Insurance Association.

Journal ArticleDOI

Journal ArticleDOI
TL;DR: Some key results involving the indifference-zone approach to selection problems for normal means and for Bernoulli “success” probabilities are discussed and possible ways of bridging the gap are proposed.
Abstract: SYNOPTIC ABSTRACTThis invited article is written in connection with the 30th anniversary of the first published papers on selection and ranking procedures. Special emphasis is placed on the indifference-zone approach. The factors leading to this author's involvement in research in this area, and the corresponding roles of other early research workers, are discussed. The philosophy of the author with respect to selection and ranking procedures is elucidated. Some key results involving the indifference-zone approach to selection problems for normal means and for Bernoulli “success” probabilities are discussed. The gap between theory and practice for selection procedures is considered, and possible ways of bridging the gap are proposed. Some unsolved problems are posed.


Journal ArticleDOI
TL;DR: A FORTRAN IV program is presented for ranking data in ascending order, and the Spearman's rank correlation coefficient, rs, as well as the corresponding t-value, can be calculated.

Journal ArticleDOI
TL;DR: A general mathematical framework for information retrieval models is presented, giving more insight into the evaluation mechanisms that may be used in IR, and a number of requirements for operators improving recall and precision are investigated.
Abstract: In this article, a general mathematical framework for information retrieval models is presented, giving more insight into the evaluation mechanisms that may be used in IR. In this framework a number of requirements for operators improving recall and precision are investigated. It is shown that these requirements can be satisfied by using descriptor weights and one combination operator only. Moreover, information items and queries—virtual items—can be treated in exactly the same way, reflecting the fact, that they both are descriptions of a number of concepts. Evaluation is performed by formulating similarity homomorphisms from query descriptions to retrieval status values that allow ranking items accordingly. For good comparisons, however, ranking must be done by the achieved proportion of perfect similarity rather than by similarity itself. In term independence models, this normalization process may satisfy the homomorphism requirement of algebra under certain conditions, but it contradicts the requirement in the term dependence case. Nevertheless, by demanding separability for the unnormalized part of the evaluation measure, it can be guaranteed that the query is evaluated to each item descriptor by descriptor, and the normalization value must be calculated only once for the whole query, using the query as a whole. Also, for each real item, the normalization value must have been calculated only once, using the item as a whole.

Proceedings ArticleDOI
05 Jun 1985
TL;DR: One of the most crucial parameters of any document retrieval system is its efficiency measured in terms of the computer resources used, particularly the time taken to retrieve output to submitted queries.
Abstract: One of the most crucial parameters of any document retrieval system is its efficiency measured in terms of the computer resources used, particularly the time taken to retrieve output to submitted queries. This quantity is especially significant for retrieval systems with large-slzed document files and/or in the case when immediate response to a user's query is required (on-llne systems, for instance). In these circumstances it is usually necessary to organize the document file so that, in order to retrieve the output to a user's query, the search is limlted to a small part of the file only.


Proceedings ArticleDOI
05 Jun 1985
TL;DR: The techniques used in ANNOD include: automatic stemming of words, common word deletion, thesaurus expansion, a complex empirical matching (ranking) algorithm (similarity measure), and techniques expressly designed to permit rapid response in a mini-computer environment.
Abstract: ANNOD is the name of a system developed at the National Library of Medicine (NLM), which implements a set of linguistic and empirical techniques that permit retrieval of natural language information in response to natural language queries. The system is based on Dr. Gerard Salton's SMART [1] document retrieval system and is presently implemented on a mini-computer as part of an Interactive TExt Management System, ITEMS.[2] Actual experience with retrieval of information from NLM's Hepatitis Knowledge Base (HKB), an encyclopedic hierarchical, full-text file, is presented. The techniques used in ANNOD include: automatic stemming of words, common word deletion, thesaurus expansion, a complex empirical matching (ranking) algorithm (similarity measure), and techniques expressly designed to permit rapid response in a mini-computer environment. Preliminary testing demonstrates high efficiency in identifying portions of a text which are relevant to users.

Journal ArticleDOI
TL;DR: On comparing the subjective ranking and the rankings attributed by FUZZIE, conclusions were made on the strengths and weaknesses of the information trace method.
Abstract: In a previous article [1] T. de Heer described an information retrieval system (FUZZIE) that uses the information trace method. Since then a statistical investigation has been made into the performance of FUZZIE. Subjects attributed a subjective ranking to answers produced by FUZZIE. On comparing the subjective ranking and the rankings attributed by FUZZIE, conclusions were made on the strengths and weaknesses of the information trace method.

01 Jan 1985
TL;DR: In this paper, the authors describe the rationale behind the ranking framework, which consists of identifying benefits, establishing a priority rank of benefits, evaluating the quality of data, and the development of a budget.
Abstract: The Gas Research Institute (GRI) used a ranking technique to select its research programs. The authors describe the rationale behind the ranking framework, which consists of identifying benefits, establishing a priority rank of benefits, evaluating the quality of data, and the development of a budget. The technique is a decision-making process that stresses cost-benefit analysis, but which does not replace expert opinion on technical matters. 7 references, 2 tables.