Frame-based Knowledge Representation Research Articles

A pre-control method in safety risk management of metro construction

ABSTRACT With the rapid development of metro construction in China, construction safety accidents occur frequently due to the lack of effective safety risk controls. Continuously improving the level of safety risk pre-control has become a significant factor in the safety managements of metro construction. This paper presents a reasoning model for safety risk pre-control measures in metro construction using case-based reasoning. To establish the case database, a frame-based knowledge representation method is adopted based on the characteristics of metro construction safety accident cases. Additionally, a two-level retrieval strategy combining hierarchical retrieval and nearest-neighbor algorithms is used for case retrieval. Furthermore, a new approach, the cloud model, is proposed to determine the attribute weight of metro construction safety accidents. This approach addresses quantitative attributes with fuzziness and randomness. Finally, a specific metro station is taken as a case to validate the feasibility of the approach. The results showed that the case-based reasoning method enables engineers to quickly and effectively develop risk pre-control measures based on previous cases. In addition, the use of case-based reasoning in decision-making enhances knowledge sharing and reuse in the metro construction industry. This paper presents an effective reasoning approach for pre-control measure analysis in metro construction, allowing engineers to quickly and efficiently carry out scientific and reasonable pre-control measures from past cases.

Semi-Supervised Event Extraction Incorporated With Topic Event Frame

Supervised Meta-event extraction suffers from two limitations: (1) The extracted meta-events only contain local semantic information and do not present the core content of the text; (2) model performance is easily degraded because of labeled samples with insufficient number and poor quality. To overcome these limitations, this study presents an approach called frame-incorporated semi-supervised topic event extraction (FISTEE), which aims to extract topic events containing global semantic information. Inspired by the frame-based knowledge representation, a topic event frame is developed to integrate multiple meta-events into a topic event. Combined with the tri-training algorithm, a strategy for selecting unlabeled samples is designed to expand the training sets, and labeling models based on conditional random field (CRF) are constructed to label meta-events. The experimental results show that the event extraction performance of FISTEE is better than supervised learning-based approaches. Furthermore, the extracted topic events can present the core content of the text.

Medical expert system for low back pain management: design issues and conflict resolution with Bayesian network.

The paper focuses on the development of a reliable medical expert system for diagnosis of low back pain (LBP) by proposing an efficient frame-based knowledge representation scheme and a suitable resolution logic with conflicts in outcomes being resolved using Bayesian network. Considering that LBP is classified into many diseases based on different pain generators, the proposed methodology infers non-conflicting LBP diseases sorted according to their chances of occurrence. A satisfactory clinical efficacy (average relative error - 0.09, recall 74.44%, precision 76.67%, accuracy 71.11%, and F1-score 73.88%) of the proposed methodology has been found after validating the design with empirically selected thirty LBP patient cases. Constraining that an inferred disease having chance of occurrence, prior to pathological investigations, below 0.75 (as set by four pain specialists) is not accepted clinically; the design can correctly identify, on average, 74.44% of actual diagnosis; and 76.67% of inferred diagnosis is included in actual diagnosis. With the predicted chance of occurrence being lower than 0.75 by a fraction of 0.09 on average, the proposed design performs well for 73.88% cases detecting 71.11% inferred outcomes as accurate. The design offers homogeneity to the actual outcomes, with the chi-squared static being calculated as 11.08 having 12 as degree of freedom. Graphical abstract.

Developing A Case-Based Reasoning Model for Safety Accident Pre-Control and Decision Making in the Construction Industry.

Case-based reasoning (CBR) has been extensively employed in various construction management areas, involving construction cost prediction, duration estimation, risk management, tendering, bidding and procurement. However, there has been a dearth of research integrating CBR with construction safety management for preventing safety accidents. This paper proposes a CBR model which focuses on case retrieval and reuse to provide safety solutions for new problems. It begins with the identification of case problem attribute and solution attribute, the state of hazard is used to describe the problem attribute based on principles of people’s unsafe behavior and objective’s unsafe state. Frame-based knowledge representation method is adopted to establish the case database from dimensions of slot, facet and facet’s value. Besides, cloud graph method is introduced to determine the attribute weight through analyzing the numerical characteristics of expectation value, entropy value and hyper entropy value. Next, thesaurus method is employed to calculate the similarity between cases including word level similarity and sentence level similarity. Principles and procedures have been provided on case revise and case retain. Finally, a real-world case is conducted to illustrate the applicability and effectiveness of the proposed model. Considering the high potential for pre-control and decision-making of construction safety accident, the proposed model is expected to contribute safety managers to take decisions on prevention measures more efficiently.

Correction to: Ontological commitments of frame-based knowledge representations

In Hommen (2017), I refer to the work of Garcia in Garcia (2015a, b). In this addendum, I would like to supplement additional references to these papers.

Ontological commitments of frame-based knowledge representations

In this paper, I shall assess the ontological commitments of frame-based methods of knowledge representation. Frames decompose concepts (statements, theories) into recursive attribute-value structures. The question is: are the attribute values in frames to be interpreted as universal properties or rather as tropes? I shall argue that universals realism and trope theory face similar complications as far as non-terminal values, i.e., values which refer to the determinable properties of objects, are concerned. It is suggested that these complications can be overcome if one is prepared to adopt an ontology of structured complexes. Such an ontology, in turn, is indifferent as to whether attribute values are interpreted as universals or as tropes.

Logical Connections of Statements at the Ontological Level

In the classical formal logics, the negation can only be applied to formulas, not to terms and predicates. In (frame-based) knowledge representation, an ontology contains descriptions of individuals, concepts and slots, that is statements about individuals, concepts and slots. The negation can be applied to slots, concepts and statements, so that the logical implication should be considered for all possible combinations of individuals, concepts, slots and statements. In this regard, the logical implication at the ontological level is different from that at the logical level. This paper attempts to give such logical implications between individuals, concepts, slots, statements and their negations.

A frame-based representation for a bedside ventilator weaning protocol

We describe the use of a frame-based knowledge representation to construct an adequately-explicit bedside clinical decision support application for ventilator weaning. The application consists of a data entry form, a knowledge base, an inference engine, and a patient database. The knowledge base contains database queries, a data dictionary, and decision frames. A frame consists of a title, a list of findings necessary to make a decision or carry out an action, and a logic or mathematical statement to determine its output. Frames for knowledge representation are advantageous because they can be created, visualized, and conceptualized as self-contained entities that correspond to accepted medical constructs. They facilitate knowledge engineering and provide understandable explanations of protocol outputs for clinicians. Our frames are elements of a hierarchical decision process. In addition to running diagnostic and therapeutic logic, frames can run database queries, make changes to the user interface, and modify computer variables.

On Pose Estimation for Human-Robot Symbiosis

This paper presents a vision based pose estimation system using knowledge based approach for human-robot symbiosis. The system is based on visual information of the face by connected component analysis of the skin color segmentation of images in HSV color model and is commenced with the face recognition and pose classification scheme using subspace PCA based pattern-matching strategies. With the knowledge of the known user's profile, face poses are then classified by multilayer perceptron. Based on the frame-based knowledge representation approach, face poses are being interpreted using the Software Platform for Agent and Knowledge (SPAK) management. On face pose recognition, robot is then instructed to perform some specific tasks by issuing pose commands. Experimental results demonstrate that the subspace method is better than that of the standard PCA method for face pose classification. The system has been demonstrated with the implementation of the algorithm to interact with an entertainment robot named, AIBO for human-robot symbiotic relationship.

Knowledge model-based adaptive intelligent control of robots for a symbiotic human robot system

This paper proposes a novel knowledge model-based adaptive intelligent control of robots for a symbiotic human-robot system. In this method, a knowledge model is first defined by the frame-based knowledge representation. It contains various frames for describing different users, features of multiple robots as well as robot behaviours for human-robot interaction and performing various tasks. According to this knowledge model, the intelligent control of robots in a symbiotic human-robot system can be implemented by means of a software platform, called Software Platform for Agents and Knowledge Management (SPAK). In addition, a kind of learning function is developed and integrated into the SPAK in order that the system can autonomously learn new knowledge by human-robot interaction and generate new control strategy for robots. Hence, the symbiotic human-robot system can adapt to various situations with different human requests and realise high-autonomous intelligent control of robots. In this paper, the effectiveness of the proposed method is verified by the experiment using actual robots.

Knowledge model-based heterogeneous multi-robot system implemented by a software platform

In order to build a heterogeneous multi-robot system that can be regarded as a primitive prototype of a future symbiotic autonomous human-robot system, this paper presents a knowledge model-based heterogeneous multi-robot system implemented by a software platform. With using frame-based knowledge representation, a knowledge model is constructed to describe the features of heterogeneous robots as well as their behaviors according to human requests. The required knowledge for constructing a heterogeneous multi-robot system can be therefore integrated in a single model and shared by robots and users. Based on the knowledge model, the heterogeneous multi-robot system is defined in the Software Platform for Agents and Knowledge Management (SPAK) by use of XML format. With the use of SPAK, the cooperative operation of heterogeneous multi-robot system can be flexibly carried out. The proposed system not only can integrate heterogeneous robots and various techniques for robots, but also can automatically perform human-robot interaction and plan robot behaviors taking into account different intelligence of robots corresponding to human requests. In this paper, an actual heterogeneous multi-robot system comprised by humanoid robots (Robovie, PINO), mobile robot (Scout) and entertainment robot dog (AIBO) is built and the effectiveness of the proposed system is verified by experiment.

An ontological approach to support the description of nursing practice in Japan with the ICNP

Background With increasing computerization of nursing records in Japan, standardization of nursing terminology is becoming imperative. Although some efforts have been made to formalize description of nursing practice in Japan with the International Classification of Nursing Practice (ICNP), lack of effective description tools has impacted negatively on the initiatives. Purpose To develop and evaluate an ontological approach that could be used to facilitate the description of nursing practice in Japan with the ICNP. Methodology An ontology-based support system was developed using Protégé-2000, mainly by the following three steps: (1) representing a standard classification of nursing practice ( the Nursing Master) in Japan; (2) representing a Japanese version of the ICNP; (3) designing an ontology-based framework. A heuristic matching algorithm was developed to automatically match the action labels in the Nursing Master with the terms of the eight axes of the ICNP Nursing Actions Classification. A preliminary evaluation was performed to examine the usefulness of the system. Results High hit rate was shown on the ICNP axes Action Type, Target, and Location. The evaluation indicated that 51.7 ± 5.8% (mean ± S.D.) of the action labels with only one action type were properly matched, and that in 80 ± 4% (mean ± S.D.) of action labels with more than one action type, at least one valid action type was matched correctly. Conclusion The ontology-based approach using a frame-based knowledge representation system (e.g., Protégé-2000) is useful for supporting the formal description of nursing practice in Japan with the ICNP.

Towards a knowledge-based scheduling system for semiconductor testing

This paper describes our efforts towards the development of a knowledge-based scheduling system for the scheduling of semiconductor testing operations. Semiconductor testing is the final phase of a four-phase semiconductor manufacturing process. The scheduling problem is a static deterministic version of what we call the workstation scheduling problem (WS). WS can be characterized as a generalized job shop problem with both parallel workstation clusters and batch processors; it requires sequencing and routeing decisions on discrete (i.e. nonbatch) and batch workstations. The proposed system consists of two distinct components: a knowledge base developed using a frame-based knowledge representation scheme, and a solution strategy based on filtered beam search. The scheduling mechanism and the associated knowledge representation scheme have been implemented on a Sun Sparc station 1+ using the programming language Common Lisp. We report on our experience to date with the system on three dimensions: computational experience, decision support capability, and knowledge management. Further research directions are suggested.

Knowledge management in software design: a tool and a trial

The paper describes experiences in applying a knowledge management tool, called CODE, to the major knowledge management problems in the design of a large commercial software system. There is explanation of how the design of such a product was documented as a CODE knowledge base by using CODE to carefully analyse all the important concepts and terms in direct consultation with the developers. CODE could be described variously as a ‘design knowledge capture tool’, a ‘conceptual design assistant’, or a ‘knowledge management system for technical documentation’. Its knowledge base and its highly interactive interface combine ideas from frame-based knowledge representation systems, object-oriented systems and hypertext systems. CODE is intended to augment and eventually replace conventional word-processing and graphical tools for the capture, editing and documentation of such knowledge. Hence it facilitates constructing definitions or descriptions of all the main concepts in a system, validating these so that the system designers are in agreement, and retrieving and graphically displaying such knowledge in various formats. CODE has been used in a wide variety of knowledge management situations, but here the focus is on its use in software engineering design.

A massively parallel knowledge-base server using a hypercube multiprocessor

In this paper we study the parallel implementation of a traditional frame based knowledge representation system for a general purpose massively parallel hypercube architecture (such as the Connection Machine CM-2). We show that, using a widely available parallel system (instead of a special purpose architecture), it is possible to provide multiple users with efficient shared access to a large scale knowledge-base. Parallel algorithms are presented for answering multiple inference, assert and retract queries on both single and multiple inheritance hierarchies. In addition to theoretical time complexity analysis, empirical results obtained from extensive testing of a prototype implementation are presented.

Intelligent operation of distribution network

The authors present an intelligent distribution control system operating in conjunction with a SCADA system with minimum supervision and interaction with a human operator. The application of a model-based expert system results in an intelligent power distribution system with learning, self organising and diagnostic capabilities. The work deviates from previous approaches in two important aspects: the use of model-based reasoning allows many different application knowledge bases (e.g. fault diagnosis, restoration, maintenance planning, etc.) to operate on the common model of the distribution network thus making it very compact and versatile, and it introduces for the first time the concept of intelligence in a power distribution system. To support modular development and simple maintenance of the expert system, the knowledge required to do the reasoning has been separated into a model of the distribution system and several knowledge bases which work on it. The frame-based object-oriented knowledge representation provides the flexibility to extend the model without adversely affecting the performance. The expert system has been tested extensively for operation with a real-time 22 kV SCADA system covering a large area of 180 substations. The expert system gives accurate fault analysis for all the fault cases and proposes the best restoration plans in less than 15 seconds. In addition, it advises the control centre operators in cases of emergency, helps them in maintenance planning and can be used as a training simulator in offline mode.

A frame-based approach to conformance testing

This paper presents the usage of the frame-based knowledge representation method for protocol specification, and shows its application in conformance testing and test sequence generation.

A knowledge-based configurator that supports sales, engineering, and manufacturing at AT&T Network Systems

PROSE is a knowledge-based configurator platform for telecommunications products. Its outstanding feature is a product knowledge base written in C-classIC, a frame-based knowledge representation system in the KL-ONE family of languages. It is one of the first successful products using a KL-ONE style language. Unlike previous configurator applications, the PROSE knowledge base is in a purely declarative form that provides developers with the ability to add knowledge quickly and consistently. The PROSE architecture is general and is not tied to any specific telecommunications product. As such, it is being reused to develop configurators for several different products. Finally, PROSE not only generates configurations from just a few high-level parameters, but it can also verify configurations produced manually by customers, engineers, or salespeople. The same product knowledge, encoded in C-classIC, supports both the generation and the verification of product configurations.

Object Representation of the Operative Environment

This paper presents a new model of knowledge representation, that will constitute a basic for the implementation of a language of representation and manipulation of knowledge in artificial vision. These concepts are inspired by different approaches based on object notion: object oriented languages, frame-based knowledge representation, actor language, and abstract type of data. This language may be a powerful tool for the resolution of artificial vision applications.

AI-based decision support tool for strategic decision making in the factory of the future

The process of strategic decision making is highly complex. Top executives make decisions based on uncertain information, much of which is difficult to quantify, often using intuitive feelings and experience as guiding factors. The demands placed upon high-level managers will increase in the factory of the future, as organizations become less hierarchical, cycle times decrease and customized production becomes a reality. Advances in process technologu will provide high levels of flexibility but knowledge technology, allied to strong and purposeful leadership, will provide the key to competitive success in many industries in future. This paper describes a prototype decision support system and learning tool which uses qualitative frame-based knowledge representation to encapsulate the intuitive knowledge of high-level decision makers. The system allows a user to construct a ‘mind set’ of a particular strategic problem domain, using qualitative relationships to describe the causal effects and interactions between concepts. The resulting knowledge base may then be used for experimentation by altering individual relationships and observing the effects of the change in the structure of the knowledge base.