Memory-based Pattern Completion in Database Semantics

2019

Information incompleteness is a major data quality issue which is amplified by the increasing amount of data collected from unreliable sources. Assessing the completeness of data is crucial for determining the quality of the data and the validity of query answers.In this work, we tackle the issue of extracting and reasoning about complete and missing information under relative information completeness setting. Under this setting, the completeness of a dataset is assessed with respect to a complete reference dataset. We advance the field by proposing two contributions: a pattern model for providing minimal covers summarizing the extent of complete and missing data partitions and a pattern algebra for deriving minimal pattern covers for query answers to analyze their validity.The completeness pattern framework presents an intriguing opportunity to achieve many applications, particularly those aiming at improving the quality of tasks impacted by missing data. Data imputation is a well-...

2012

Semantic search on databases often return a large number of results, only a small subset of which is relevant to the user. We present a semantic search technique considering the type of desired Web resources and the semantic relationships between the resources and the query keywords Ranking and categorization, which can also be combined, to alleviate this information overload problem. Results refinement for databases is the focus of this work. A novel search interface that enables the user to navigate large number of query results using the Pattern recognition. First, the query results are matched with the key word using full Pattern matching. In contrast, previous works expand the hierarchy in a predefined static manner, without navigation refinement modeling. We show that the problem of selecting the best concepts to reveal at each refinement and propose an efficient Pattern matching algorithm. We show experimentally that how results are refined using full Pattern matching at firs...

2003

Data intensive applications produce complex information that is posing requirements for novel Database Management Systems (DBMSs). Such information is characterized by its huge volume of data and by its diversity and complexity, since the data processing methods such as pattern recognition, data mining and knowledge extraction result in knowledge artifacts like clusters, association rules, decision trees and others. These artifacts, called patterns, need to be stored and retrieved effectively and efficiently. In this paper, we review the concept of patterns and their applicability in several research domains and we define the knowledge domain related to the PANDA project. We examine the different types of patterns that are extracted from a data set, in order to gather the necessary requirements for the definition of a pattern model. This model will constitute the heart of the Pattern Base Management System that will be designed.

Advances in computational intelligence and robotics book series, 2017

Patterns are mentioned usually in the extraction context. Little stress is posed in their representation and management. This chapter is focused on the representation of the patterns, manipulation with patterns and query patterns. Crucial issue can be seen in systematic approach to pattern management and specific pattern query language which takes into consideration semantics of patterns. In the background we discuss two different approaches to the pattern store and manipulation (based on inductive database and PANDA project). General pattern model is illustrated using abstract data type implemented in Oracle. In the following chapters the introduction to querying patterns and simple scheme of the architecture PBMS is shown.

Proceedings of the 2005 International Conference on Intelligence Analysis, Mitre Corporation and Office of the Assistant Director of Central Intelligence for Analysis and Production, 2-4 May 2005, McLean, VA, 2005

This paper describes some fundamental issues associated with using rule-based reasoning to discover evidence for the instantiation of complex threat types in relational data. Based on the considerations raised, I delineate a general model for the pattern recovery process.

Proceedings of the second ACM symposium on Symbolic and algebraic manipulation - SYMSAC '71, 1971

Information and Computation/information and Control, 1987

hci20, 2020

Database Semantics (DBS) models the cycle of natural language communication as a transition from the hear to the think to the speak and back to the hear mode (turn taking). In contradistinction to the substitution-driven sign-based approaches of truth-conditional semantics and phrase structure grammar, DBS is data-driven and agent-based. The purpose is a theory of semantics for an autonomous robot with language. Propositions are content in DBS, instead of denoting truth values (Sects. 1-3). Content is built from the semantic kinds of referent, property, and relation , which are concatenated by the classical semantic relations of structure , i.e. functor-argument and coordination. To enable reference as an agent-internal cognitive process, language and nonlanguage contents use the same computational data structure and operation kinds, and differ mostly in the presence vs. absence of language-dependent surface values. DBS consists of (i) an interface, (ii) a memory, and (iii) an operation component. 1 The interface component mediates between the agent's cog-nition and its external and internal environment, represented as raw data provided by sensors and activators (Sects. 4-7). The data of the agent's moment by moment monitoring are stored at the memory's now front. As part of the on-board control unit, the now front is the location for performing the procedures of the operation component, resulting in content. keywords: data structure, data base schema, pattern matching, turn taking, type-token, grounding, sensory and processing media and modalities, reference 1 Building Content in the Agent's Hear Mode DBS defines a content in terms of concepts like square (7.1) or blue (7.3) connected with the classical semantic relations of structure such as subject/predicate and conjunct−conjunct. The concepts are supplied by the agent's memory and defined as types. In recognition, they are activated by matching raw data provided by the interface component, resulting in tokens. 2 In action, a type is adapted to a current purpose as a token and realized as raw data (7.2, 7.4). For concatenation, the concepts are embedded as core values into nonrecur-sive feature structures with ordered attributes, called proplets. Proplets serve as the computational data structure of DBS. The semantic relations between proplets are established by address. Consider the following example: 1 The components correspond roughly to those of a von Neumann machine (Neumann, J.v., 1945): the (i) interface component corresponds to the vNm input-output device, the (ii) memory (database) component corresponds to the vNm memory, and the (iii) operation component performs functions of the vNm arithmetic-logic unit. 2 The type-token distinction was introduced by C. S. Peirce (CP 4:537).

2004

Most results (lemmas and theorems) providing conditions under which associative memories are able to perfectly recall patterns of a fundamental set are very restrictive in most practical applications. In this note we describe a simple but effective procedure to transform a fundamental set of patterns (FSP) to a canonical form that fulfils the propositions. This way pattern recall is strongly improved. We provide numerical and real examples to reinforce the proposal.

Lecture Notes in Computer Science, 2021

The transfer of information processed by human beings from their short-term memory (STM) to their semantic memory creates two kinds of knowledge: a semantic network of associations and a structured set of rules to govern human deliberate behaviour under explicit attention. This paper focuses on the memory processes that create the first of these two kinds of knowledge. Human memory storage and processing are modeled using the Real-time Maude rewrite language. Maude's capability of specifying complex data structures as many sorted algebras and the time features of Real-Time Maude are exploited for (1) providing a means for formalising alternative memory models, (2) modelling in silico experiments to compare and validate such models. We aim at using our model for the comparison of alternative cognitive hypothesis and theories and the analysis of interactive systems.