We present a probabilistic relational algebra (PRA) which is a generalization of standard relational algebra. Here tuples are assigned probabilistic weights giving the probability that a tuple belongs to a relation. Based on intensional semantics, the tuple weights of the result of a PRA expression always confirm to the underlying probabilistic model. We also show for which expressions extensional semantics yields the same results. Furthermore, we discuss complexity issues and indicate possibilities for optimization. With regard to databases, the approach allows for representing imprecise attribute values, whereas for information retrieval, probabilistic document indexing and probabilistic search term weighting can be modelled. As an important extension, we introduce the concept of vague predicates which yields a probabilistic weight instead of a Boolean value, thus allowing for queries with vague selection conditions. So PRA implements uncertainty and vagueness in combination with the...

In this paper, a probabilistic relational model is presented which combines relational algebra with probabilistic retrieval. Based on certain independence assumptions, the operators of the relational algebra are redefined such that the probabilistic algebra is a generalization of the standard relational algebra. Furthermore, a special join operator implementing probabilistic retrieval is proposed. When applied to typical document databases, queries can not only ask for documents, but for any kind of object in the database. In addition, an implicit ranking of these objects is provided in case the query relates to probabilistic indexing or uses the probabilistic join operator. The proposed algebra is intended as a standard interface to combined database and IR systems, as a basis for implementing user-friendly interfaces. 1 Introduction The fields of databases (DB) and information retrieval (IR) have been coexisting for a very long time, but with little influence on each other. IR peop...

Abstract: Integration of information retrieval (IR) in database management systems (DBMSs) has proven difficult. Previous attempts to integration suffered from inherent performance problems, or lacked desirable separation between logical and physical data models. To overcome these problems, we discuss a database approach based on structural object-orientation. We implement IR techniques using extensions in an object algebra called MOA. MOA has been implemented on top of the database backend Monet, a state-of-the-art highperformance database kernel with a binary relational interface. Our prototype implementation of the inference network retrieval model using MOA and Monet demonstrates the feasibility of this approach. We conclude with a discussion of the advantages of our database design.

The integration of information retrieval (IR) and database systems requires a data model which allows for modelling documents as entities, representing uncertainty and vagueness and performing uncertain inference. For this purpose, we present a probabilistic data model based on relations in nonfirst -normal-form (NF2). Here, tuples are assigned probabilistic weights giving the probability that a tuple belongs to a relation. Thus, the set of weighted index terms of a document are represented as a probabilistic subrelation. In a similar way, imprecise attribute values are modelled as a set-valued attribute. We redefine the relational operators for this type of relations such that the result of each operator is again a probabilistic NF2 relation, where the weight of a tuple gives the probability that this tuple belongs to the result. By ordering the tuples according to decreasing probabilities, the model yields a ranking of answers like in most IR models. This effect also can be used for ...

In this paper, we present a document model which integrates the logical structure and hypertext link structure of hyperdocuments in order to manage structured documents with hypertext links. Based on this model we define a new structure query language which expresses the structure query using path expressions. To process a structure query in a document management system which represents structure information as database relations, costly join operations are used to find a relationship between elements in a document hierarchy. In order to overcome this problem, schemes based on the parse tree [6] and element locator [2] have been used. In this paper, we propose a new structure query processing scheme that uses unique element identifiers (UID's) to evaluate structure queries. Our scheme has advantage over previous schemes since it can obtain the UID's of the ancestors and descendents directly from the UID of a node without disk access. We present relational database schemas for our schem...

Modern fact databases contain abundant data classified through several classifications.

Modern extensible database systems support multimedia data through ADTs. However, because of the problems with multimedia query formulation, this support is not sufficient. Multimedia querying requires an iterative search process involving many different representations of the objects in the database. The support that is needed is very similar to the processes in information retrieval. Based on this observation, we develop the miR Ror architecture for multimedia query processing. We design a layered framework based on information retrieval techniques, to provide a usable query interface to the multimedia database. First, we introduce a concept layer to enable reasoning over low-level concepts in the database. Second, we add an evidential reasoning layer as an intermediate between the user and the concept layer. Third, we add the functionality to process the users ’ relevance feedback. We then adapt the inference network model from text retrieval to an evidential reasoning model for multimedia query processing. We conclude with an outline for implementation of miR Ror on top of the Monet extensible database system. 1

. Applications which require a combination of structured data with unstructured text fields are becoming of increasing practical interest. But whereas structured data are usually stored in a relational database, large text collections are maintained by proprietary text or information retrieval systems. The synthesis of both areas is still a topic of intensive research. We describe one such application, namely maintaining library catalogues, and study the efficiency of two implementation alternatives both based on RDBMS technology. In the first alternative word occurrence information is encoded using bitlists. The other chooses a direct implementation within the relational model. Performance tests are done which are based on real world data and real world user transactions. They demonstrate that the problem of the bitlist implementation is caused by conversions which are necessary to combine them with structured data. In contrast, our direct implementation benefits from today's sophisti...

Abstract not found

. Applications which require a combination of structured data with unstructured text fields are becoming of increasing practical interest. But whereas structured data are usually stored in a relational database, large text collections are maintained by proprietary text or information retrieval systems. The synthesis of both areas is still a topic of intensive research. We describe one such application, namely maintaining library catalogues, and study the efficiency of two implementation alternatives both based on RDBMS technology. In the first alternative word occurrence information is encoded using bitlists. The other chooses a direct implementation within the relational model. Performance tests are done which are based on real world data and real world user transactions. They demonstrate that the problem of the bitlist implementation is caused by conversions which are necessary to combine them with structured data. In contrast, our direct implementation benefits from today's sophisti...