Automatic Knowledge Acquisition Research Articles

Stanford University's AI in medicine: still cutting the edge

Since the early 1970s, Stanford University Medical School has been not only one of the leading medical schools, but also home to some of the best researchers in artificial intelligence. The author discusses automated knowledge acquisition and representation, protocol directed therapy, time dependent data mining and medical decision support systems.

A heuristic search algorithm for path determination with learning

We present and analyze an algorithm, adaptive A*(AA*), for finding a least-cost-path from start node to goal node set in a directed graph. Arc costs are assumed to be scalar-valued, and the cost of each path is the sum of the concomitant arc costs. Search is guided by: 1) a collection of real-valued functions on the node set, which is a generalization of the heuristic function associated with A*; 2) a set of predetermined optimal paths; and 3) a set of paths in the graph that are considered desirable but may or may not be optimal. The knowledge representations described in (1) and (3) can be useful in describing knowledge acquired from humans. The knowledge representation described in (2) can be used to automate knowledge acquisition, so that A* exhibits a form of machine learning. Additionally, the collection of real-valued functions on the node set can be useful in describing bounds on the perfect heuristic function, i.e., the solution of the related dynamic program. A numerical analysis, using a specialization of AA* applied to a model of the Cleveland, OH, road network demonstrated significant performance improvement relative to A*.

FILM: a fuzzy inductive learning method for automated knowledge acquisition

Inductive learning that creates decision trees from a set of existing cases is useful for automated knowledge acquisition. Most of the existing methods in literature are based on crisp concepts that are weak in handling marginal cases. In this paper, we present a fuzzy inductive learning method that integrates the fuzzy set theory into the regular inductive learning processes. The method converts a decision tree induced from regular method into a fuzzy decision tree in which hurdle values for splitting branches and classes associated with leaves are fuzzy. Results from empirical tests indicate that the new fuzzy approach outperforms the popular discriminant analysis and ID3 in predictive accuracy.

Are artificial neural networks black boxes?

Artificial neural networks are efficient computing models which have shown their strengths in solving hard problems in artificial intelligence. They have also been shown to be universal approximators. Notwithstanding, one of the major criticisms is their being black boxes, since no satisfactory explanation of their behavior has been offered. In this paper, we provide such an interpretation of neural networks so that they will no longer be seen as black boxes. This is stated after establishing the equality between a certain class of neural nets and fuzzy rule-based systems. This interpretation is built with fuzzy rules using a new fuzzy logic operator which is defined after introducing the concept of f-duality. In addition, this interpretation offers an automated knowledge acquisition procedure.

An efficient algorithm for automatic knowledge acquisition

This paper presents a new algorithm for extracting IF-THEN rules from examples. The algorithm employs an efficient rule searching method and a simple metric for assessing rule generality and accuracy. The paper illustrates step by step the operation of the algorithm and discusses its performance on the IRIS data classification problem.

Providing understanding of the behavior of feedforward neural networks

The advent of artificial neural networks has stirred the imagination of many in the field of knowledge acquisition. There is an expectation that neural networks will play an important role in automating knowledge acquisition and encoding, however, the problem solving knowledge of a neural network is represented at a subsymbolic level and hence is very difficult for a human user to comprehend. One way to provide an understanding of the behavior of neural networks is to extract their problem solving knowledge in terms of rules that can be provided to users. Several papers which propose extracting rules from feedforward neural networks can be found in the literature, however, these approaches can only deal with networks with binary inputs. Furthermore, certain approaches lack theoretical support and their usefulness and effectiveness are debatable. Upon carefully analyzing these approaches, we propose a method to extract fuzzy rules from networks with continuous-valued inputs. The method was tested using a real-life problem (decision-making by pilots involving combat situations) and found to be effective.

Some Techniques of Reducing the Dangers of Combinatorial Explosion in Automatic Knowledge Acquisition

The danger of getting into a combinatorial explosion has remained probably the most serious impediment to using knowledge-based systems for real-life problems. Justifiable heuristics must cut down the size of potentially huge search spaces to a realistic and manageable size - without jeopardizing the success of finding satisfactory solutions. This paper describes some techniques that have proved to be effective in the operation of two different systems: the Quasi-Optimizer (QO) and the Generalized Production Rule System (GPRS). The QO is a domain-independent automatic knowledge acquisition tool that can obtain, verify, fuse and optimize human expertise. It generates computer models, descriptive theories, of human decision making strategies, and can also select and combine the best components of several such models into a single strategy which may be considered a normative theory in the statistical sense. The techniques developed to reduce the danger of combinatorial explosion with the QO include selecting the most independent (near-orthogonal) decision variables, chunking conceptually coherent decision variables, eliminating statistical outlier values, and using dynamically evolving experimental designs to result in a near-uniform reliability distribution of the strategy responses. The GPRS can estimate/predict values of hidden variables and can thus serve as a module of an expert system in need of numerical or functional estimates of hidden variable values. (Hidden variables can be observed and measured only intermittently and at irregular points of time and space - in contrast with open variables whose values can be identified at any time and location.) The estimation is based on generalized production rules expressing stochastic and potentially causal relations between known values of hidden variables and certain mathematical properties of the open variable distribution. A multi-dimensional learning process adds to, consolidates and optimizes the generalized rule base. It gradually merges “similar enough” production rules, and eliminates spurious and statistically not justifiable ones. Such processes reduce the danger of combinatorial explosions in knowledge acquisition. Finally, we note that human decision makers appear to resort to similar methods, although in a less systematic manner, when they are confronted with very large decision spaces.

Constructability Analysis: Machine Learning Approach

Computerized constructability analysis, using knowledge-based tools, is the key to effective construction process automation. However, the development of such tools requires a prior acquisition of constructability knowledge. In this paper, results of a feasibility study of automated constructability knowledge acquisition are reported. In the conducted research, constructability of a beam in a reinforced-concrete frame has been investigated. For this problem, a knowledge representation space has been developed, relevant constructability data has been acquired from industry, and a collection of examples has been prepared. In this collection, each example represents a structural design concept evaluated from the point of view of its constructability. The collection of examples has been used by a learning system to acquire from them constructability knowledge in the form of decision rules. The experience gained while conducting the automated knowledge acquisition process has been used to develop initial methodological conclusions regarding the use of learning systems in constructability analysis and to determine the future research directions.

Modified fractal signature (MFS): a new approach to document analysis for automatic knowledge acquisition

One of the key technologies related to knowledge and data engineering is the acquisition of knowledge and data in the development and utilization of information system and the strategies to capture new knowledge and data. Actually, millions of documents, including technical reports, government files, newspapers, books, magazines, letters, bank checks, etc., have to be processed every day, and knowledge has to be acquired from them. This paper presents a new approach to document analysis for automatic knowledge acquisition. The traditional approaches have two major disadvantages: (1) They are not effective for processing documents with high geometrical complexity. Specially, the top-down approach can process only the simple documents which have specific format or contain some a priori information. (2) The top-down approach needs to split large components into small ones iteratively, while the bottom-up approach needs to merge small components into large ones iteratively. They are time consuming. This new approach is based on modified fractal signature. It can overcome the above weaknesses.

Deep hydrodesulfurization of atmospheric gas oil; Effects of operating conditions and modelling by artificial neural network techniques

Artificial neural networks (ANN) are currently being explored in various engineering fields as valuable tools for automatic model-building and knowledge acquisition. This technique was applied to model hydrodesulfurization of atmospheric gas oil in a mini-pilot trickle-bed reactor. Sulfur removal was measured as a function of temperature, pressure and liquid hourly space velocity (LHSV) for three sulfur feed concentrations. The potential of a two-stage process was also tested. A set of experimental data was used to teach a three-layer neural network. The capability of the artificial neural network to predict the performance was tested with a different set of data. The agreement between predicted and experimental values was good. Temperature, LHSV and staging of the process were determined to be important parameters, while pressure had a little effect over the range tested in this study.

Learning engineering: The key to automatic knowledge acquisition

AbstractLearning engineering is a new subarea of knowledge engineering dealing with the methodological aspects of using learning systems in knowledge acquisition. In this paper, the justification for the development of Learning Engineering is provided, and its major subdomains and research directions are briefly discussed.

Role-limiting methods for automated knowledge acquisition: A problem-solving perspective

In this short paper we provide some complementary remarks on an excellent paper published earlier in this journal. The issues discussed here are concerned with similarities and differences between knowledge acquisition (KA) in expert systems on the one hand and requirement acquisition (RA) in other kinds of information systems (particularly decision support systems, or DSSs) on the other hand, We focus on role-limiting methods developed for automated knowledge acquisition, examining these methods and their integration from the problem solving perspective. We also suggest that a feasibility study on the potential of developing similar methods for the tasks of RA may benefit DSS and other kinds of information systems.

Role-limiting methods for automated knowledge acquisition: a problem-solving perspective

Role-limiting methods for automated knowledge acquisition: a problem-solving perspective

Automated fuzzy knowledge acquisition with connectionist adaptation

This paper presents an automated knowledge acquisition architecture for the truck docking problem. The architecture consists of a neural network block, a fuzzy rule generation block and a genetic optimisation block. The neural network block is used to quickly and adaptively learn from trials the driving knowledge. The fuzzy rule generation block then extracts the driving knowledge to form a knowledge rule base. The driving knowledge rule base is further optimised in the genetic optimisation block using a genetic algorithm. Computer simulations are presented to show the effectiveness of the architecture.

Evaluation of automatic knowledge acquisition techniques in the diagnosis of acute abdominal pain

Clinical diagnosis in acute abdominal pain is still a major problem. Computer-aided diagnosis offers some help; however, existing systems still produce high error rates. We therefore tested machine learning techniques in order to improve standard statistical systems. The investigation was based on a prospective clinical database with 1254 cases, 46 diagnostic parameters and 15 diagnoses. Independence Bayes and the automatic rule induction techniques ID3, NewId, PRISM, CN2, C4.5 and ITRULE were trained with 839 cases and separately tested on 415 cases. No major differences in overall accuracy were observed (43–48%), except for Newld, which was below the average. Between the different techniques some similarities were found, but also considerable differences with respect to specific diagnoses. Machine learning techniques did not improve the results of the standard model Independence Bayes. Problem dimensionality, sample size and model complexity are major factors influencing diagnostic accuracy in computer-aided diagnosis of acute abdominal pain.

Toward the ultimate synthesis/recognition system.

This paper predicts speech synthesis, speech recognition, and speaker recognition technology for the year 2001, and it describes the most important research problems to be solved in order to arrive at these ultimate synthesis and recognition systems. The problems for speech synthesis include natural and intelligible voice production, prosody control based on meaning, capability of controlling synthesized voice quality and choosing individual speaking style, multilingual and multidialectal synthesis, choice of application-oriented speaking styles, capability of adding emotion, and synthesis from concepts. The problems for speech recognition include robust recognition against speech variations, adaptation/normalization to variations due to environmental conditions and speakers, automatic knowledge acquisition for acoustic and linguistic modeling, spontaneous speech recognition, naturalness and ease of human-machine interaction, and recognition of emotion. The problems for speaker recognition are similar to those for speech recognition. The research topics related to all these techniques include the use of articulatory and perceptual constraints and evaluation methods for measuring the quality of technology and systems.

PRIMEROSE: PROBABILISTIC RULE INDUCTION METHOD BASED ON ROUGH SETS AND RESAMPLING METHODS

Automated knowledge acquisition is an important research issue in machine learning. Several methods of inductive learning, such as ID3 family and AQ family, have been applied to discover meaningful knowledge from large databases and their usefulness is assured in several aspects. However, since their methods are of a deterministic nature and the reliability of acquired knowledge is not evaluated statistically, these methods are ineffective when applied to domains essentially probabilistic in nature, such as medical domains. Extending concepts of rough set theory to a probabilistic domain, we introduce a new approach to knowledge acquisition, which induces probabilistic rules based on rough set theory (PRIMEROSE) and develop a program that extracts rules for an expert system from a clinical database, using this method. The results show that the derived rules almost correspond to those of the medical experts.

Towards an automatic consensus generator tool: EGAC

Automatic knowledge acquisition for expert systems has attracted much attention. Several algorithms that infer concept descriptions from a given set of training examples have been developed to aid in this task, some of them elicit concepts from examples organized in data matrices. These algorithms infer from different training examples (or different matrices defined by a set of experts) slightly different concept descriptions. Herein we propose a method based on synthesis of judgements, fuzzy sets and classification methods that applied to a set of data matrices builds an agreed one that synthesises the information contained in the set of matrices. The method proposed can be applied to data matrices with attributes of several types: measure and ratio quantitative attributes, and ordered and nonordered qualitative ones. >

Automated knowledge acquisition by reasoning failures

Automated knowledge acquisition is viewed as a problem in modeling a knowledge engineer's introspective capabilities. We formulate a computational model of automated knowledge acquisition by modeling such introspective debugging actions. We propose that an automated knowledge acquisition system should be provided with an explicit model of performance-failure explanation mechanisms, and show that linking the expectations of the knowledge-based system to the model enables the knowledge acquisition program to determine what parts of the domain knowledge base are responsible for observed performance failures. The knowledge acquisition process is failure driven and is guided by the explanations of failures. Generating explanations of bugs in the knowledge base is perceived as the abductive as well as model-based process. We use an explanation apprentice that analyzes the faulty behavior of the knowledge-based system and answers a broad range of questions of the knowledge acquisition system. A training-and-test experiment using the knowledge acquisition system increases the performance of the knowledge-based system of the training and the testing set by 31% and 51%, respectively.

Automating knowledge acquisition: a propositional approach to representing expertise as an alternative to repertory grid technique

Repertory grid technique plays a central role in the elicitation methodology of many well-reported knowledge acquisition tools or workbenches. However, the dependability of these systems is low where the technique breaks down or proves inadequate due to limited expressive power and other problems. The paper introduces an alternate approach based on Personal Construct Theory that elicits an expert's knowledge as a network of terms that constitutes a propositional formalism. An extended example is used to both highlight the difficulties encountered using repertory grids and illustrate how these are overcome using the proposed approach. The results of an empirical study are presented where an experienced clinician compared the knowledge structures that she constructed for a diagnostic task using each elicitation technique. Furthermore, although the network representation is amenable to inductive learning methods for generating production rules, an inference method is demonstrated which reveals the formalism's categorical reasoning potential. The authors conclude that it is more appropriate to classify such methods as either mediating or immediate rather than the knowledge structures they employ. The paper contributes to a better understanding of constructivist formalisms developed for knowledge acquisition. >