We study an entity search/match problem that requires retrieved tuples match to an input entity query. We assume the input queries are of the same type as the tuples in a materialised relational table. Existing keyword search over relational databases focuses on assembling tuples from a variety of relational tables in order to respond to a keyword query. The entity queries in this work differ from the keyword queries in two ways: (i) an entity query roughly refers to an entity that contains a number of attribute values, i.e. a product entity or an address entity; (ii) there might be redundant or incorrect information in the entity queries that could lead to misinterpretations of the queries. In this paper, we propose a transformation that first converts an unstructured entity query into a multi-valued structured query, and two retrieval methods are proposed to generate a set of candidate tuples from the database. The retrieval methods essentially formulate SQL queries against the database given the multi-valued structured query. The results of a comprehensive evaluation of a large-scale database (more than 29 millions tuples) and two real-world datasets showed that our methods have a good trade-off between generating correct candidates and the retrieval time compared to baseline approaches.

Probabilistic record linkage is a well established topic in the literature. Fellegi–Sunter probabilistic record linkage and its enhanced versions are commonly used methods, which calculate match and non-match weights for each pair of records. Bayesian network classifiers–naive Bayes classifier and TAN have also been successfully used here. Recently, an extended version of TAN (called ETAN) has been developed and proved superior in classification accuracy to conventional TAN. However, no previous work has applied ETAN to record linkage and investigated the benefits of using naturally existing hierarchical feature level information and parsed fields of the datasets. In this work, we extend the naive Bayes classifier with such hierarchical feature level information. Finally we illustrate the benefits of our method over previously proposed methods on four datasets in terms of the linkage performance ( F_1 score). We also show the results can be further improved by evaluating the benefit provided by additionally parsing the fields of these datasets.

Automatic semantic annotation of data from databases or the Web is an important pre-process for data cleansing and record linkage. It can be used to resolve the problem of imperfect field alignment in a database or identify comparable fields for matching records from multiple sources. The annotation process is not trivial because data values may be noisy, such as abbreviations, variations or misspellings. In particular, overlapping features usually exist in a lexicon-based approach. In this work, we present a probabilistic address parser based on linear-chain conditional random fields (CRFs), which allow more expressive token-level features compared to hidden Markov models (HMMs). In additions, we also proposed two general enhancement techniques to improve the performance. One is taking original semi-structure of the data into account. Another is post-processing of the output sequences of the parser by combining its conditional probability and a score function, which is based on a learned stochastic regular grammar (SRG) that captures segment-level dependencies. Experiments were conducted by comparing the CRF parser to a HMM parser and a semi-Markov CRF parser in two real-world datasets. The CRF parser out-performed the HMM parser and the semi-Markov CRF in both datasets in terms of classification accuracy. Leveraging the structure of the data and combining the linear-chain CRF with the SRG further improved the parser to achieve an accuracy of 97% on a postal dataset and 96% on a company dataset.

Probabilistic record linkage has been well investigated in recent years. The Fellegi-Sunter probabilistic record linkage and its enhanced version are commonly used methods, which calculate match and non-match weights for each pair of corresponding fields of record-pairs. Bayesian network classifiers – naive Bayes classifier and TAN have also been successfully used here. Very recently, an extended version of TAN (called ETAN) has been developed and proved superior in classification accuracy to conventional TAN. However, no previous work has applied ETAN in record linkage and investigated the benefits of using a naturally existing hierarchy feature level information. In this work, we extend the naive Bayes classifier with such information. Finally we apply all the methods to four datasets and estimate the $$F_1$$ scores.

Mark Bishop PhD studied Cybernetics and Computer Science at the University of Reading. He is Professor of Cognitive Computing at Goldsmiths, University of London and between 2010-2014 was Chair of the society for the study of Artificial Intelligence and the Simulation of Behaviour (AISB), the largest Artificial Intelligence Society in the United Kingdom. He has published widely in areas of Artificial Intelligence, Machine Learning and Neural Computing.

There are several similar, but not identical, definitions of control dependence in the literature. These definitions are given in terms of control flow graphs which have had extra restrictions imposed (for example, end-reachability). We define two new generalisations of non-termination insensitive and non-termination sensitive control dependence called weak and strong control-closure. These are defined for all finite directed graphs, not just control flow graphs and are hence allow control dependence to be applied to a wider class of program structures than before. We investigate all previous forms of control dependence in the literature and prove that, for the restricted graphs for which each is defined, vertex sets are closed under each if and only if they are either weakly or strongly control-closed. Low polynomial-time algorithms for producing minimal weakly and strongly control-closed sets over generalised control flow graphs are given. This paper is the first to define an underlying semantics for control dependence: we define two relations between graphs called weak and strong projections, and prove that the graph induced by a set of vertices is a weak/strong projection of the original if and only if the set is weakly/strongly control-closed. Thus, all previous forms of control dependence also satisfy our semantics. Weak and strong projections, therefore, precisely capture the essence of control dependence in our generalisations and all the previous, more restricted forms. More fundamentally, these semantics can be thought of as correctness criteria for future definitions of control dependence.

We introduce a new non-strict semantics for a simple while language. We demonstrate that this semantics allows us to give a denotational definition of variable dependence and neededness, which is consistent with program slicing. Unlike other semantics used in variable dependence, our semantics is substitutive. We prove that our semantics is preserved by traditional slicing algorithms. (C) 2007 Published by Elsevier Inc.

A program schema defines a class of programs, all of which have identical statement structure, but whose expressions may differ. We define a class of syntactic similarity binary relations between linear structured schemas, which characterise schema equivalence for structured schemas that are linear, free and liberal. In this paper we report that similarity implies equivalence for linear schemas, and that a near-converse holds for schemas that are linear, free and liberal. We also show that the similarity of two linear schemas is polynomial-time decidable. Our main result considerably extends the class of program schemas for which equivalence is known to be decidable, and suggests that linearity is a constraint worthy of further investigation.

A formal relationship between program slicing and partial evaluation is established. It is proved that for terminating programs, a residual program produced by partial evaluation is semantically equivalent to a conditioned slice.

Program slicing is an automated source code extraction technique that has been applied to a number of problems including testing, debugging, maintenance, reverse engineering, program comprehension, reuse and program integration. In all these applications the size of the slice is crucial; the smaller the better. It is known that statement minimal slices are not computable, but the question of dataflow minimal slicing has remained open since Weiser posed it in 1979. This paper proves that static slicing algorithms produce dataflow minimal end slices for programs which can be represented as schemas which are free and liberal.

Program conditioning consists of identifying and removing a set of statements which cannot be executed when a condition of interest holds at some point in a program. It has been applied to problems in maintenance, testing, re-use and re-engineering. All current approaches to program conditioning rely upon both symbolic execution and reasoning about symbolic predicates. The reasoning can be performed by a ‘heavy duty’ theorem prover but this may impose unrealistic performance constraints. This paper reports on a lightweight approach to theorem proving using the FermaT Simplify decision procedure. This is used as a component to ConSUS, a program conditioning system for the Wide Spectrum Language WSL. The paper describes the symbolic execution algorithm used by ConSUS, which prunes as it conditions. The paper also provides empirical evidence that conditioning produces a significant reduction in program size and, although exponential in the worst case, the conditioning system has low degree polynomial behaviour in many cases, thereby making it scalable to unit level applications of program conditioning.

Several approaches to reverse and re-engineering are based upon program slicing. Unfortunately, for large systems, such as those which typically form the subject of reverse engineering activities, the space and time requirements of slicing can be a barrier to successful application. Faced with this problem, several authors have found it helpful to merge control flow graph (CFG) nodes, thereby improving the space and time requirements of standard slicing algorithms. The node-merging process essentially creates a 'coarser' version of the original CFG. The paper introduces a theory for defining control flow graph node coarsening calculi. The theory formalizes properties of interest, when coarsening is used as a precursor to program slicing. The theory is illustrated with a case study of a coarsening calculus, which is proved to have the desired properties of sharpness and consistency.

Th paper shows how analysis of programs in terms of pre- and postconditions can be improved using a generalisation of conditioned program slicing called pre/post conditioned slicing. Such conditions play an important role in program comprehension, reuse, verification and reengineering. Fully automated analysis is impossible because of the inherent undecidability of pre- and post- conditions. The method presented reformulates the problem to circumvent this. The reformulation is constructed so that programs which respect the pre- and post-conditions applied to them have empty slices. For those which do not respect the conditions, the slice contains statements which could potentially break the conditions. This separates the automatable part of the analysis from the human analysis