scispace - formally typeset
Search or ask a question

Showing papers on "Ranking (information retrieval) published in 1982"


Journal ArticleDOI
TL;DR: Algorithms are presented that run in time which depends nontrivially on the rank k of the element to be selected and which is sublinear with respect to set cardinality, and all bounds are shown to be optimal to within a constant multiplicative factor.

265 citations


Journal ArticleDOI
TL;DR: In this paper, the authors considered the use of imperfect information for risk sharing and incentive purposes when perfect observation of actions and outcomes is impossible, making complete contracting infeasible, and formulated a generalized agency model to analyse this problem.
Abstract: This paper considers the use of imperfect information for risk sharing and incentive purposes when perfect observation of actions and outcomes is impossible, making complete contracting infeasible. The incentive-insurance problem is defined to consist of two parts: the choice of an information system and the design of a sharing rule based on the information system. A generalized agency model is formulated to analyse this problem. The agency models of Ross (1973a, b), Wilson (1968), Stiglitz (1974), Mirrlees (1976), Harris and Raviv (1979), Holmstrom (1979) a.o. appear as special cases of the generalized model. The analysis focuses on the value of information in the agency information problem. The set of information systems which are valuable—i.e. improve risk sharing and incentives in a Pareto sense—is characterized. A problem-independent ranking of information systems for the agency information problem is then characterized under the assumption that the agent's preferences are additive in money and actions. The ranking may be viewed as a generalization of Blackwell's ranking of information systems for decision problems, to this particular game. When the agent's risk preferences depend on his choice of action, on the other hand, it is shown that the Blackwell ranking may be invalid. Randomized incentive schemes are shown to be efficient when the incentive effect of risk is positive and sufficiently large relative to the absolute risk aversion of the partners.

193 citations


Journal ArticleDOI
TL;DR: In this paper, the Borda-Kendall method does not perform as claimed in the case of ties and a "minimum variance" method for determining the consensus ranking is proposed and its properties examined.
Abstract: This paper investigates the Borda-Kendall method for the determination of a consensus ranking. It is shown that in the case of ties the method does not perform as claimed. A "minimum variance" method for determining the consensus ranking is proposed and its properties examined. It is shown to be equivalent to the Borda-Kendall method if ties are not allowed. An algorithm to determine the "minimum variance" consensus ranking in the case of ties is described. Results obtained from the solution of problems of various sizes are discussed.

138 citations




Journal ArticleDOI
TL;DR: It is perhaps unnecessary to be able to determine the meaning in the absolute sense; it may be sufficient to measure how far there is a conformity in meaning, and then only in the context of the set of documents in which the answer to a query is sought.
Abstract: The success of information retrieval depends on the ability to measure the effective relationship between a query and its response. If both are posed in natural language, one might expect that understanding the meaning of that language could not be avoided. The aim of this research is to demonstrate that it is perhaps unnecessary to be able to determine the meaning in the absolute sense; it may be sufficient to measure how far there is a conformity in meaning, and then only in the context of the set of documents in which the answer to a query is sought. Handling a particular language using a computer is made possible through replacing certain texts by special sets. A given text has a ‘syntactic trace’, the set of all the overlapping trigrams forming part of the text. When determining the effective relationship between a query and its answer, not only do their syntactic traces play a role, but so do the traces of all other documents in the set. This is known as the ‘information trace method’.

39 citations



Journal ArticleDOI
TL;DR: In this article, the authors consider a choice problem among alternative space allocations, and employ the methodology of multi-attribute value functions to score and rank seven suggested area assignments, through explicit consideration of trade-offs among the four space categories, yielding a rank ordering which is consistent with the decision maker's preferences.
Abstract: The paper considers a choice problem among alternative space allocations. The choice is often made intuitively because of the difficulty in simultaneously considering multiple criteria (i.e., space categories), some of which are of conflicting nature. The paper employs the methodology of multi-attribute value functions to score and rank seven suggested area assignments. Through explicit consideration of trade-offs among the four space categories, the methodology yields a rank ordering which is consistent with the decision maker's preferences. In addition to ranking the alternatives, the formal analysis provided important insight into the trade-off and value structure over the space attributes to facilitate a new area assignment which was superior to the ones originally proposed.

27 citations


Proceedings ArticleDOI
George E. Heidorn1
16 Jun 1982
TL;DR: In this article, a metric that can be easily computed during either bottom-up or top-down construction of a parse tree for ranking the desirability of alternative parses is presented.
Abstract: This brief paper, which is itself an extended abstract for a forthcoming paper, describes a metric that can be easily computed during either bottom-up or top-down construction of a parse tree for ranking the desirability of alternative parses. In its simplest form, the metric tends to prefer trees in which constituents are pushed as far down as possible, but by appropriate modification of a constant in the formula other behavior can be obtained also. This paper includes in introduction to the EPISTLE system being developed at IBM Research and a discussion of the results of using this metric with that system.

22 citations


Journal ArticleDOI
TL;DR: This paper considers the generating, ranking, and unranking of 2-3 trees with n keys and proposes a linear ordering among these trees that can be done in time after a preprocessing step that takes time and space.
Abstract: In this paper we consider the generating, ranking, and unranking of 2-3 trees with n keys. We propose a linear ordering among these trees. The problem of ranking is to determine the rank of a given tree in this ordering, while unranking means constructing the tree of a given rank. The main result is that ranking and unranking can be done in $O(n)$ time after a preprocessing step that takes $O(n^2 )$ time and space.

12 citations


Patent
08 Oct 1982
TL;DR: In this paper, the arrangement and retrieval of many document files with easy operation in efficient way, by executing the designation with attribute table formed in hierarchy and retrieval processing with tree structure retrieval file.
Abstract: PURPOSE:To realize the arrangement and retrieval of many document files with easy operation in efficient way, by executing the designation with attribute table formed in hierarchy and retrieval processing with tree structure retrieval file. CONSTITUTION:Document managing data comprising an attribute managing table and attribute data stored on a recording medium is transferred to a system control section 3 so as to organize the retrieval data file of the tree structure by each document attribute and the file is registered in a retrieval data file section 6. A check item A1 is designated by looking into a document attribute display A of list form of the highest rank from an operation display section 4, the item is retrieved by using the tree structure file of the retrieval file section 6 and the document attribute B of the list form of the next ranking is displayed.

Journal ArticleDOI
Gerard Salton1
01 Sep 1982
TL;DR: A new Boolean retrieval environment is outlined in which the queries are automatically constructed from the original natural language query formulations provided by the users, and which produce better retrieval output than conventional retrieval operations based on manually prepared query statements.
Abstract: Conventional information retrieval systems use Boolean query formulations and inverted file technologies for search and retrieval purposes The need to construct complex Boolean queries in order to obtain the benefit of the existing retrieval operations constitutes a substantial burden for the users In most environments trained search intermediaries are used to facilitate the communication between system and userIn this note a new Boolean retrieval environment is outlined in which the queries are automatically constructed from the original natural language query formulations provided by the users Any available Boolean query formulations can also be improved automatically by using the natural language text of previously retrieved documents identified as relevant during previous searches The automatic queries can be formulated in a standard Boolean system, or in an extended system in which the interpretation of the Boolean operators and and or is relaxed In either case the automatic Boolean manipulations produce better retrieval output than conventional retrieval operations based on manually prepared query statements



Journal ArticleDOI
TL;DR: In this paper, the authors show that existing procedures are vitiated by interaction in such a design, but give a new procedure which is fully valid even in the presence of unbounded interaction.
Abstract: : Over the years, while it has been recognized that developments for other designs are of great interest, almost all ranking an selection papers have dealt explicitly only with the completely randomized design. We review what is known in other design settings, then proceed to new results on the complete factorial experimental design. We show that existing procedures are vitiated by interaction in such a design, but give a new procedure which is fully valid even in the presence of unbounded interaction. (Author)

01 May 1982
TL;DR: The methods are based on well-founded ranking and are applicable to both just and fair computations and do not assume a decrease of the rank at each computation step.
Abstract: : In this paper, one of a series on verification of concurrent programs the authors present proof methods for establishing eventuality and until properties. The methods are based on well-founded ranking and are applicable to both just and fair computations. These methods do not assume a decrease of the rank at each computation step. It is sufficient that there exists one process which decrease the rank when activated. Fairness then ensures that the program will eventually attain its goal. In the finite state case the proofs can be represented by diagrams. Several examples are given. (Author)

Journal ArticleDOI
TL;DR: Several statistical sampling methods are evaluated for estimating the total number of relevant documents in a collection for a given query, where a smooth curve is developed which relates precision to document rank.
Abstract: Several statistical sampling methods are evaluated for estimating the total number of relevant documents in a collection for a given query. The total number of relevant documents is needed in order to compute recall values for use in evaluating document retrieval systems. The simplest method considered uses simple random sampling to estimate the number of relevant documents. Another type of random sampling, which assigns unequal selection probabilities to the individual documents in the collection, is also investigated. An alternative approach considered uses curve fitting and extrapolation, where a smooth curve is developed which relates precision to document rank. Another curve relates a function of precision to the query-document score. In either case, the curve is extrapolated to the total number of documents in order to estimate the number of relevant documents. Empirical comparisons are made of all three methods.


Journal ArticleDOI
TL;DR: A method using the amount of semantic information of query terms as weight in a fuzzy relation of resemblance can be used to partially order documents in decreasing order of resemblance with the query.
Abstract: A method using the amount of semantic information of query terms as weight in a fuzzy relation of resemblance is presented. The relation can be used to partially order documents in decreasing order of resemblance with the query. Large operational bibliographic data bases are used to test the validity of the approach.

Patent
02 Feb 1982
TL;DR: In this article, a transfer ranking control system for transferring a data between multiple channel control devices and a main memory device is proposed, where transfer priority is determined by transfer ranking of a subchannel.
Abstract: PURPOSE:To reduce a data transfer waiting time and to increase a data transfer factor, by deciding transfer priority ranking by transfer ranking of a subchannel, in a transfer ranking control system for transferring a data between plural channel control devices and a main memory device. CONSTITUTION:Transfer request controlling circuits A14, B15 are provided on channel control devices A8, B12, respectively, and in case when a transfer request from a high speed input/output control device 3 connected with the channel control device B12 whose transfer priority ranking is low has occurred, a transfer request from a low speed input/output control device 2 connected with the channel control device A8 whose transfer priority ranking is high is suppressed, a transfer request from the high speed input/output device 3 connected with the channel control device B12 whose transfer priority ranking is low is made effective, and a data is transferred to a main memory device 13.

Patent
24 Jun 1982
TL;DR: In this article, a refueling machine selective button is provided on card readers CR1, CR2 of a service station, and a user pushes its button and inserts a card into a card insertion port.
Abstract: PURPOSE:To process a processing request in order of received applications by setting receipt ranking in response to a request for coupling the second data to the first data of the processing requests generated at about the same time, and holding an area on the memory in accordance with the receipt ranking irrespective of generated ranking of the second data. CONSTITUTION:A refueling machine selective button is provided on card readers CR1, CR2 of a service station, and a user pushes its button and inserts a card into a card insertion port. A data which has been read by the respective readers CR1, CR2 from said card is sent to an output device D, and by the device D, whether said data is proper or not is decided and the received ranking is decided. Also, a signal for starting and finishing the refueling from refueling machines P1-P3, and information of the quantity of refueling are sent to the output device D. Subsequently, the receipt ranking of processing requests is set by the output device D, an area on the memory decided by the receipt ranking is held irrespective of the ranking of the signal from the refueling machines P1-P3, the first request of the residual received ranking is coupled with the refueling machines P1-P3, and refueling is performed in order or received applications.

Proceedings ArticleDOI
18 May 1982
TL;DR: A specific IR model was developed which allows to deduct suitable query and ranking evaluation algorithms and it is shown that weighted query terms occurring in the query formulation may be weighted.
Abstract: A data organization for information retrieval (IR) systems is described which uses the structures imposed on the stored information. Trees are used as the main structure of data as information contents are often hierarchically structured (e.g. classifications, thesauri). However, these trees have been expanded to pseudo networks by so-called cross connecting paths. So-called data connecting paths link the information structures and the main data file. Terms occurring in the query formulation may be weighted. These weights are interpreted and then used by both the retrieval and ranking algorithm. One of the paramount problems is how to combine weighted query terms. Since the well-known IR schemes (Boolean retrieval, fuzzy retrieval etc.) do not work in our environment, a specific IR model was developed which allows to deduct suitable query and ranking evaluation algorithms.

Patent
09 Feb 1982
TL;DR: In this paper, a symbol set input consisting of a prescribed symbol is inputted to operating circuits 1-5, the number of black of the prescribed mesh is caught as a feature, and a stable coefficient of said feature is calculated.
Abstract: PURPOSE:To effectively extract a feature, to effectively make a dictionary after selecting an optimum feature value rank, and to reduce the capacity of the dictionary, by appreciating the ranking of plural features which have been prepared in advance, in accordance with a set unit. CONSTITUTION:A symbol set input consisting of a prescribed symbol is inputted to operating circuits 1-5, the number of black of a prescribed mesh is caught as a feature, and a stable coefficient of said feature is calculated. In said each operating circuit 1-5, a stable coefficient of each different feature is calculated, and each stable coefficient is supplied to a comparing circuit 6. In the comparing circuit 6, ranking is executed in order from a small stable coefficient of each feature, it is outputted, and a feature is extracted effectively. In this way, ranking is executed to an optimum feature value, an effective dictionary is made, and a symbol is recognized easily.

Journal ArticleDOI
TL;DR: Possibility of the program creating virtual world models to prevent the user unnecessarily accessing the whole data base, as well as incorporating a form of weighting based on association strenghs between items in the data base are explored.
Abstract: Elsewhere, there is a description of a computer program design which aims at satisfying incompletely defined user needs through a man-machine dialogue which does not require a query formulation by the user. The program attempts to satisfy the user’s needs from a model which it builds of his interests a model which is a subset of its own ’world model’, the data base. This paper discusses current work based on this program design. Possibilities of the program creating virtual world models to prevent the user unnecessarily accessing the whole data base, as well as incorporating a form of weighting based on association strenghs between items in the data base are explored.



Journal ArticleDOI
TL;DR: In this paper, the squared sums of the differences between the observed and calculated chemical shift values are assumed to be proportional to the dissimilarity of the real and possible structures of the structures.