Abstract: The paper presents a review of the ONIONS project. ONIONS is committed to developing a large-scale, axiomatized ontology library for medical terminology. The developed methodology exploits a description logic-based design for the modules in the library and makes extended use of generic theories, thus creating a stratification of the modules. Terminological knowledge is acquired by conceptual analysis and ontology integration over a set of authoritative sources. After addressing general issues about conceptual analysis and integration, the methodology is briefly described. The central part of the paper presents the investigation we have made on the 476,000 medical concepts singled out by the National Library of Medicine as the Metathesaurus TM in the UMLS project. This is followed by several case studies concerning lexical polysemy, the interface between ontologies. A section describing the current structure of the library and the generic theories reused is provided. Current results of our research include the integration of some top-level ontologies in the ON9.2 ontology library, and the formalization of the terminological knowledge in the UMLS Metathesaurus.

Abstract: A Data Warehouse (DW) is a database that collects and stores data from multiple remote and heterogeneous information sources. When a query is posed, it is evaluated locally, without accessing the original information sources. In this paper we deal with the issue of designing a DW, in the context of the relational model, by selecting a set of views to materialize in the DW. First, we briefly present a theoretical framework for the DW design problem, which concerns the selection of a set of views that (a) fit in the space allocated to the DW, (b) answer all the queries of interest, and (c) minimize the total query evaluation and view maintenance cost. We then formalize the DW design problem as a state space search problem by taking into account multiquery optimization over the maintenance queries (i.e., queries that compute changes to the materialized views) and the use of auxiliary views for reducing the view maintenance cost. Finally, incremental algorithms and heuristics for pruning the search space are presented.

Abstract: Most private and public organizations have recently turned their attention to the process by which they operate, to improve service and product quality and customer satisfaction. To support business process reengineering, methods and tools for process modeling and analysis are required. The paper presents the ARTEMIS methodology and associated tool environment for business process analysis for reengineering. In the ARTEMIS methodological framework, business processes are modeled as workflows and are analyzed according to an organizational structure perspective and an operational structure perspective. With these two perspectives, the analyst can plan reengineering interventions based on the degree of autonomy/dependency of organization units in terms of coupling, and the inter-process semantic correspondences, in terms of data and operation similarity, respectively. The ARTEMIS methodology and associated tool environment have been conceived and applied in the framework of the PROGRESS research project. In the paper, we report on a reengineering case study of this project involving the Italian Ministry of Justice.

Abstract: Techniques that allow a program to generate E-R models for small application domains are described. The inputs to the program are natural language specifications. The program consists of two major components: a natural language understander and an E-R generator. The paper describes briefly the knowledge representation structures constructed by the natural language understander, and, then, explains the E-R generator in detail. The E-R generator consists of two kinds of rules: specific rules linked to the semantics of some words in the sentences, and generic rules that identify ER-entities and ER-relationships on the basis of the logical form of the sentence and on the basis of the ER-entities and ER-relationships under construction. The program has been tested on a set of database problems.

Abstract: Subject-based classification is an important part of information retrieval, and has a long history in libraries, where a subject taxonomy was used to determine the location of books on the shelves. We have been studying the notion of subject itself, in order to determine a formal ontology of subjects for a large-scale digital library card catalog system. Deep analysis reveals a lot of ambiguity regarding the usage of subjects in existing systems and terminology, and we attempt to formalize these notions into a single framework for representing it.

Abstract: Some database models have already been developed to deal with complex values but they have constrains that data stored is precise and queries are crisp. However, as many researchers have pointed out, there is a need to present, manipulate, and query complex and uncertain data of various non-traditional database applications such as oceanography, multimedia, meteorology, office automation systems, engineering designs, expert database systems and geographic information systems. In this paper, we present a logical database model, which is an extension of a nested relational data model (also known as an NF 2 data model), for representing and manipulating complex and uncertain data in databases. We also introduce a possible physical representation of such complex and uncertain values in databases and describe the query processing of the model that we discuss here.

Abstract: The relational data model has become the standard for mainstream database processing despite its well-known weakness in the area of representing application semantics. The research community's response to this situation has been the development of a collection of semantic data models that allow more of the meaning of information to be presented in a database. The primary tool for accomplishing this has been the use of various data abstractions, most commonly: inclusion, aggregation and association. This paper develops a general model for analyzing data abstractions, and then applies it to these three best-known abstractions.

Abstract: Task models and problem solving methods can be specified informally or formally. In recent years various approaches have formalized the notion of task models or problem solving methods. Most modelling approaches concentrate on the form of a task model or problem solving method rather than on their precise semantics: a formalisation is often only a syntactical formalisation. A more precise definition of the semantics requires explication of the control of a system's behaviour. In this paper temporal semantics is defined for a compositional modelling approach to task models and problem solving methods. The semantics is a description of a compositional system's behaviour; a temporal approach provides a means to describe the dynamics involved. The formalisation of the semantics is based on compositional three-valued temporal models. The compositional structure of information states, transitions and reasoning traces provides a transparent model of the system's behaviour, both conceptually and formally.

Abstract: In any particular combination of domains, although a common understanding of the underlying low-level concepts concerning a domain attribute may exist, different concept hierarchies may have been built at different data-holding sites. A distributed database may therefore hold different views of the same data, or differently classified samples from the same population. Because of this, when statistical functions are applied to generate summary tables, the resulting summary-based partitions may be heterogeneous. In these situations, integration of such summary-based partitions can reveal latent information at a new, and finer, level of granularity. In this paper, the classification schemes are described using a matrix representation of the intersection hypergraph, and efficient numerical algorithms are proposed to determine the optimal granularity of the integrated summary data.

Abstract: In a mobile environment, the broadcast paradigm has been recognized as an effective way to disseminate frequently requested information to multiple ‘mobile clients’. In this paper, we show that to query the broadcast database efficiently, database items must be organized over the broadcast channel in a meaningful way, reflecting the semantics of the database being broadcast. We define, for the first time, a cost model called the Semantic Ordering Model for identifying the optimal organization of database items over the broadcast channel. Experimental results are also delineated to illustrate the feasibility of our cost model, using relational and object-oriented databases as examples.

Abstract: In this paper, we propose a formal model for multilevel databases This model aims at being a generic model, that is it can be interpreted for any kind of database (relational, object-oriented) Our model has three layers The first layer corresponds to a model for a non-protected database The second layer corresponds to a model for a multilevel database In this second layer, we propose a list of theorems that must be respected in order to build a secure multilevel database We also propose a new solution to manage cover stories without using the ambiguous technique of polyinstantiation The third layer corresponds to a model for a MultiView database, that is, a database that provides at each security level a consistent view of the multilevel database Finally, as an illustration, we interpret our 3-layer model in the case of an object-oriented database

Abstract: Many organizations nowadays own several information systems storing precious data whose effective exploitation is a key issue in many contexts. A major problem to be faced is the semantic normalization of scheme objects used in heterogeneous, independent and pre-existing databases as to single out differences and similitudes among data whereby a consistent, integrated view of available information can be obtained. To solve large instances of this problem, the development of automatic support tools appears to be mandatory. This paper gives a contribution in this framework by presenting algorithms for extracting useful properties holding among objects belonging to sets of database schemes. The algorithms are capable of deriving both nominal and structural properties of scheme objects starting from database scheme descriptions. Some application examples are presented to demonstrate the effectiveness of our approach.

Abstract: We present an approach to handle incremental materialization of object-oriented views. Queries that define views are implemented as methods that are invoked to compute corresponding views. To avoid computation from scratch each time a view is accessed, we introduce some deferred update algorithms that reflect for a view only related modifications introduced into the database while that view was inactive. A view is updated by considering modifications performed within all classes along the inheritance and class-composition subhierarchies rooted at every class used in deriving that view. To each class, we add a modification list to keep one modification tuple per view dependent on that class. Such a tuple acts as a reference point that marks the start of the next update to the corresponding view.

Abstract: Human knowledge in any expertise area changes with respect to time. Two types of such knowledge can be identified, time independent and time dependent . It is shown that the maintenance effort of the latter is harder than that of the former. The present paper applies research results in the area of temporal databases, in order to maintain a rule-based knowledge base whose content changes with respect to the real world time. It is shown that the approach simplifies the maintenance of time dependent knowledge. It also enables the study of the evolution of knowledge with respect to time, which is knowledge on its own . Three distinct solutions are actually proposed and evaluated. Their common characteristic is that knowledge is stored in a database; therefore, all the advantages of databases are inherited by knowledge bases. Implementations are also reported.

Abstract: The main motivation for the development of a query algebra of complex objects QAL is the study of the operations needed to query the structural aspects of object-oriented databases which are due to the advantages of the object-oriented database model over conventional relational and functional database models. The basic operations of the QAL query algebra evolved from relational algebras and the FQL family of functional query languages by refining their operations for the manipulation of objects. In order to support the features specific to object-oriented data models, QAL offers: (i) a set of operations which provide the means for querying conceptual schemata and (ii) an operation which provides a simple and efficient way of querying nested components of complex objects. We show through a case-study consisting of a comprehensive set of examples how these operations can be employed to express a class of queries specific to object-oriented databases

Abstract: Case-based reasoning (CBR) systems define knowledge in terms of a memory or library of past cases and a retrieval mechanism that revolves around retrieving data relevant to a goal query. Additionally, such systems employ an adaptation component that transforms the retrieved data into a solution to the problem expressed by the original query. The combination of goal query and the subsequent solution transformation is referred to as CBR goal query. Goal queries are concerned with data that is close to the request expressed in the query. Conventional relational and object-oriented databases are usually concerned specific queries. Extending conventional object-oriented data models, this paper proposes a concept-oriented data model that provides a variety of mechanisms to support conventional goal and CBR goal queries. It is shown that such a concept-oriented data model could be used as the core for a more general knowledge base management system.

Abstract: The presentation of multimedia data is not only characterized by precise temporal constraints; also spatial constraints must be taken into account. An important requirement in multimedia systems is thus the integrated modeling of spatio-temporal constraints. Moreover, it is important to devise methods for checking the consistency of the specified constraints. In this paper, we first propose a spatio-temporal object graph (STOG) model that provides an integrated and graphical representation of spatio-temporal constraints. Second, we investigate consistency conditions between spatial and temporal constraints expressed in the STOG model. Then, we present a prototype system implementing the proposed model.

Abstract: A data warehouse collects and maintains a large amount of data from several distributed and heterogeneous data sources. Often the data is stored in the form of materialized views in order to provide fast access to the integrated data, regardless of the availability of the data sources. In this paper we focus on the following problem: for a given set of materialized select-project-join (SPJ) views, how can we find and minimize the auxiliary data stored in a data warehouse in order to make all materialized views in the data warehouse self-maintainable? For this problem we first devise an algorithm for finding such an auxiliary view set by exploiting information sharing among the auxiliary views and materialized views themselves to reduce the total size of auxiliary views. We then consider how to make the data warehouse still self-maintainable by minor modifications when there is a view addition to or deletion from it by giving an algorithm for this incremental maintenance purpose.

Abstract: Methods for the reuse of specification knowledge have been developed to reduce the costs of systems development. One approach is to build libraries of reusable analysis patterns, i.e. models describing the generic features of a situation that can occur in different domains. In order to systematise libraries of such patterns, we propose a novel analysis pattern based on a deontic perspective. The basic components of this pattern are object types representing obligations, the parties involved in these obligations, their respective roles, and the speech acts that create and terminate the obligations. We argue that this pattern captures specification knowledge at an appropriate level of abstraction, has wide applicability, and effectively supports designers in the construction of models. Furthermore, a number of instances of this pattern are analysed and classified into different categories.

Abstract: In an integrated dissemination-based information system (DBIS), data are disseminated to clients in two modes. In the first mode, data publication mode, frequently accessed data are periodically pushed (published) to clients, who only need to listen to the communication channel and filter the incoming data stream for the desired objects. The second mode, demand-driven mode, employs a request-response mechanism for clients to explicitly request for data that are not published. For an integrated DBIS to be effective, the two modes of information dissemination must be carefully employed. For example, publishing too many data objects can lead to excessively long access time for the publication mode even when the workload (i.e., number of clients) is low. On the other hand, publishing very few data objects can lead to excessive waiting for clients using the demand-driven mode when the communication channel becomes a bottleneck. In this paper, we identify the kinds of bottleneck that may arise in an integrated DBIS, and propose a novel algorithm to tune such a system for “optimal” performance (in terms of average access time). We also study the effect of different feedback mechanisms that can help the server to adapt to changing access patterns. We evaluated the various proposed schemes based on a simulation model, and report our findings in this paper.

Abstract: This paper proposes a hybrid classification learning system for databases that integrates rule induction and lazy learning. For rule induction learning, we use an entropy function based on Hellinger divergence to measure the amount of information each inductive rule contains. For lazy learning, we also use the Hellinger measure to automatically generate attribute weights and to compute similarities between data values of non-numeric data types. Our system has been implemented and tested extensively on a number of well-known machine learning data sets. The performance of our system was favorable compared to those of other well-known classification learning techniques based on monostrategic learning methods.

Abstract: Known methods for checking integrity constraints in deductive databases do not eliminate all aspects of redundancy in integrity checking. By making the redundancy aspects of integrity constraint checking explicit, independently from any chosen method, it is possible to develop a new method that is optimal with respect to the classified redundancy aspects. We distinguish three types of redundancy and propose an integrity checking method based on revised inconsistency rules.

Abstract: We propose a new replication control scheme for mobile database systems. Replicating data inevitably induces the burden of maintaining consistency, which requires more complex synchronization mechanisms, especially in the presence of communication failures. If a communication link fails, all sites are divided into two disjoint groups: the major group called the quorum partition and the minor group called the non-quorum partition. This phenomenon, called network partitioning, could lead to access starvation. In order to access replicated databases without access starvation even in the presence of network partitioning, we propose a new scheme called fault-tolerant quorum consensus (FTQC). FTQC is based on the idea that quorum formation is allowed even in the non-quorum partition. FTQC guarantees inter-partition one-copy serializability, without sacrificing data availability.

Abstract: Serializability requires that the execution of each transaction must give the illusion to be an atomic action, i.e., the execution of a set of transactions must appear to be a serial one. This requirement, however, is too strong and unnecessarily restricts concurrency among transactions, when semantic information is available for the transaction processing mechanism. In this paper, a new correctness criterion for concurrent execution of database transactions, denoted semantic serializability, is proposed. Semantic serializability is based on the use of semantic information about database objects (and not about transactions). The main idea of our proposal is to provide different atomicity views for each transaction and, for this reason, to allow interleavings among transactions which are nonserializable, but which preserve database consistency. We develop two concurrency control protocols, which are based on semantic serializability. One protocol is based on a locking-mechanism and the other one uses a non-locking approach. Our proposal is suitable to a wide variety of advanced database applications, such as CAx, MDBS, GIS and WFMS.

Abstract: In a wireless computing environment, a server disseminates information by periodically broadcasting data on ‘air’, while clients ‘catch’ their desired data on the fly. To determine the data to be broadcast, the server constructs a broadcast program. While a nonuniform broadcast program has been demonstrated to be effective in reducing the average access times for single-record retrievals, existing nonuniform broadcast programs perform poorly for range queries. In this paper, we propose a new algorithm to generate nonuniform broadcast programs that can facilitate range queries. Our algorithm supports selective tuning using the flexible indexing scheme. We conducted an extensive simulation study, and our results show that the proposed algorithm can generate programs that lead to significant improvement (in terms of both access time and tuning time) for range queries without sacrificing much on the performance of single object retrievals.

Abstract: Despite the fact that transaction implementation in most commercial database management systems uses locking for concurrency control, optimistic concurrency control has recently gained attention for its efficiency in new types of data-intensive applications such as computer-aided design, computer-aided software engineering, and computer integrated manufacturing applications. This paper describes and analyzes a new concurrency control algorithm referred to as precise serialization, a variation of an optimistic concurrency control algorithm using forward validation that is known to outperform other optimistic algorithms proposed so far. We show that the forward validation-based algorithm incurs transaction restarts which are not absolutely necessary to ensure data consistency, and that the precise serialization algorithm solves this problem with a reasonably limited amount of effort by dynamically adjusting transaction execution history. We illustrate, through simulation, that by removing such unnecessary restarts, the precise serialization algorithm improves the performance of the forward validation-based algorithm under certain conditions.

Abstract: Modeling real-world applications typically consists of two parts: the representations of entities of interest, and the representations of relationships among the entities. Object-oriented data models support the first task. This paper argues through examples that the second modeling task is not well supported in current object-oriented database systems. An extended object-oriented data model that facilitates both modeling tasks is developed. This extension provides support for both the representation of complex relationships among entities and their manipulations (queries, updates). A prototype implemented on the ONTOS database management system is briefly described.

Abstract: Cyclic covers represent the complex relationships between predicates lying within, and outside of, a perfect model of an indefinite stratified deductive database, and are shown to facilitate the solution to a variety of problems under the perfect model semantics. Cyclic covers are shown to characterise view updates, and this yields a top–down method for solving the view insertion problem via the computation of partially cyclic covers and a simple closure operation. Weakly cyclic covers and cyclic trees are shown to yield methods of partially evaluating (i.e., pre-processing) the intensional database to facilitate query compilation, the identification of redundant rules, the removal of recursion and the removal of positive or negative intensional sub-goals. Such pre-processing is immune to future changes in the extensional database. The notion of a conjunctive answer to a database query is introduced. Such answers do not suffer from the problems of redundancy as seen with (the more commonly studied) “disjunctive” answers. In addition, conjunctive answers yield more information about their query than do their disjunctive counterparts, and in particular the set of disjunctive answers can be computed from the set of conjunctive answers, whereas the converse does not hold. Cyclic covers are shown to characterise conjunctive answers, thus yielding a top–down method of computing such answers. The problem of computing conjunctive answers is shown to be a special case of the view update problem.

Abstract: The Time Split-B Tree is an efficient storage and access structure proposed by Lomet and Salzberg for transaction-time temporal data. This structure forces strong data duplication, due to storage space splits along the time dimension, to improve selectivity for temporal queries. However, redundancy increases storage space requirements and, as a consequence, can be considered as a weakness for a large class of transaction-time database applications. From this viewpoint, almost all the merit figures of the Time Split B-Tree suffer from the high redundancy degree usually reached by the structure during its lifetime. An improvement of the Time Split B-Tree is presented in this work. It is based on a new adaptive split policy which aims to dynamically limit data duplication within given bounds. An evaluation of the Time Split B-Tree with the adaptive split policy has been done via extensive simulations in comparison with the previous Time Split B-Tree performance. The Time Split B-Tree with the new split policy proved to have a better and more uniform behaviour.