Second generation expert systems

Second generation expert systems are able to combine heuristic reasoning based on rules, with deep reasoning based on a model of the problem domain. This solves a number of major problems of current expert systems, most importantly the problem of knowledge acquisition: second generation expert systems can learn new rules by examining the results of deep reasoning. The paper outlines the components of second generation expert systems and gives an example.     

HARNESS: a next generation distributed virtual machine

Heterogeneous Adaptable Reconfigurable Networked SystemS (HARNESS) is an experimental metacomputing system [L. Smarr, C.E. Catlett, Communications of the ACM 35 (6) (1992) 45–52] built around the services of a highly customizable and reconfigurable Distributed Virtual Machine (DVM). The successful experience of the HARNESS design team with the Parallel Virtual Machine (PVM) project has taught us both the features which make the DVM model so valuable to parallel programmers and the limitations imposed by the PVM design. HARNESS seeks to remove some of those limitations by taking a totally different approach to creating and modifying a DVM.     

An embeddable virtual machine for state space generation

The semantics of modelling languages are not always specified in a precise and formal way, and their rather complex underlying models make it a non-trivial exercise to reuse them in newly developed tools. We report on experiments with a virtual machine-based approach for state space generation. The virtual machine’s (VM) byte-code language is straightforwardly implementable, facilitates reuse and makes it an adequate target for translation of higher-level languages like the SPIN model checker’s Promela, or even C. As added value, it provides efficiently executable operational semantics for modelling languages. Several tools have been built around the VM implementation we developed, to evaluate the benefits of the proposed approach.     

Lexical choice in text generation and machine translation

We present in this article the state of the art in lexical choice research in text generation and machine translation. The existing implementations are discussed with respect to four topics: (i) the place of lexical choice in the overall generation process, (ii) the information flow within the generation process and the consequences thereof for lexical choice, (iii) the internal organization of the lexical choice process, (iv) the phenomena covered by lexical choice. Possible future directions in lexical choice research are identified.     

Towards meaning-based machine translation: using abstractions from text generation for preserving meaning

One of the primary motivations of text generation is the achievement of a very wide range of linguistic abilities coupled with functional control of that range. This control rests on the appropriate construction of abstract specifications of meaning that can guide the generation process to produce language that is textually, grammatically, and lexically appropriate. Such abstract semantic specifications, when constructed in the right way, preserve much of the meaning required in a translation without unduly constraining syntactic form. This is potentially of great value for machine translation since it opens up the possibility of domain-independent, constrained, meaning-based translation. This paper describes how the upper model of the PENMAN text generation system provides a level of semantic abstraction of this kind. It offers examples of the motivation of broader sets of likely translational equivalents than that possible with transfers at lower-levels of abstraction and sets out types of constraints by which the set of likely translational equivalents may be reduced to high-quality renderings of the source text.     

Relational-grammar-based generation in the JETS Japanese-English machine translation system

This paper describes the design and function of the English generation phase in JETS, a minimal transfer, Japanese-English machine translation system that is based on the linguistic framework of relational grammar. To facilitate the development of relational grammar generators, we have built a generator shell that provides a high-level relational grammar rule-writing language and is independent of both the natural language and the application. The implemented English generator (called GENIE) maps abstract canonical structures, representing the basic predicate-argument structures of sentences, into well-formed English sentences via a two-stage plan-and-execute design. The modularity inherent in the plan-and-execute design permits the development of a very general and stable deterministic execution grammar. Another major feature of the GENIE generator is that it iscategory-driven, i.e., planning rules and execution rules are distributed over a part-of-speech hierarchy (down to individual lexical items) and are invoked via an inheritance mechanism only if appropriate for the category being processed. Categorydriven processing facilitates the handling of exceptions. The use of a syntactic planner and category-driven processing together provide a great deal of flexibility without sacrificing determinism in the generation process.     

Non-deterministic and emotional chatting machine: learning emotional conversation generation using conditional variational autoencoders

Neural Computing and Applications - Conversational responses are non-trivial for artificial conversational agents. Artificial responses should not only be meaningful and plausible, but should also...     

Second generation pleated pneumatic artificial muscle and its robotic applications

This paper reports on the second generation of the pleated pneumatic artificial muscle (PPAM) which has been developed to extend the lifespan of its first prototype. This type of artificial muscle was developed to overcome dry friction and material deformation which is present in the widely used McKibben muscle. The essence of the PPAM is its pleated membrane structure which enables the muscle to work at low pressures and at large contractions. There is growing interest in this kind of actuation for robotics applications due to its high power to weight ratio and adaptable compliance, especially for legged locomotion and robot applications in direct contact with a human. This paper describes the design of the second-generation PPAM, for which specifically the membrane layout has been changed. In terms of this new layout the mathematical model, developed for the first prototype, has been reformulated. This paper gives an elaborate discussion on this mathematical model that represents the force generation and enclosed muscle volume. Static load tests on some real muscles, which have been carried out in order to validate the mathematical model, are then discussed. Furthermore, two robotic applications are given which successfully use these pneumatic artificial muscles. One is the biped Lucy and the another one is a manipulator application which works in direct contact with an operator.     

TRIZ: Inventive Creativity Based On The Laws of Systems Development

Different problem solving methodologies drawing on various philosophical and practical approaches are in use all over the world. This paper gives details of the TRIZ method, created by G. A. Altshuller in the former Soviet Union in 1946. TRIZ uses systems thinking concepts to structure problem solving and encourages the user to discover the objective laws of evolution and development in the field of inventive creativity. Examples of the use of the system and reflections on its impact are given by the author.     

Problem-Driven Inventing: a Concept for Strong Solutions to Inventive Tasks

This article presents the Concept of Problem-Driven Inventing. It combines the application of condensed technical knowledge and special techniques for directed creativity and aims at strong solutions to inventive tasks. One core element of this concept is to work out the inherent contradiction of a company’s problem and then take it as a challenge to overcome this contradiction instead of accepting a trade-off of contradictory requirements. To achieve this one should orientate oneself by the area of conflict between the metaphors of the ideal machine on the one hand and the fitting to economical, technological and social trends on the other hand. Several strong tools for overcoming technical and scientific contradictions complete this concept and can be supported by traditional creativity techniques. The Concept of Problem-Driven Inventing combines various approaches: It is inspired by Altshuller’s empirical research work of the last six decades and uses the tools of his ”Theory of Inventive Problem Solving“ as well as other tools and techniques. Those tools are integrated into the modular structure of the ”Five-Field-Analysis“ by Müller-Merbach. Thus, this concept puts findings of different research fields in a form that is easy to understand and which is promising to be applicable to different types of enterprises in today’s economic environment as well as to individuals. The article illustrates the Concept of Problem-Driven Inventing by a continuous example, gives advice for its application and outlines current projects.     

Further analysis of code generation for a single register machine

Summary This paper presents a systematic development of the analysis of code generation for the single register machine, from linear to exponential time algorithms. A class of expressions with common subexpressions is introduced which can be optimally coded in quadratic time. A heuristic based on this development is shown to produce code size better than 5/4 that of an optimal coding, for any expression. A method is proposed which would improve this bound, but the most straightforward implementation is shown to require exponential time. Finally, a new analysis of the exponential time bound is presented which more closely describes the complexity of the task of producing optimal code.     

Inventive Tool Use to Comprehend Big Code

Software developers often need to understand a large body of unfamiliar code with little or no documentation, no experts to consult, and little time to do it. A post appeared in January 2008 on Slashdot, a technology news Web site, asking for tools and techniques that could help. This article analyzes 301 often passionate and sometimes articulate responses to this query, including the themes and the associated tool recommendations. The most common suggestions were to use a code navigation tool, use a design recovery tool, use a debugger to step through the code, create a runtime trace, use problem-based learning, ask people for help, study the code from top down, and print out all the code. This analysis presents an intriguing snapshot of how software developers in industry go about comprehending big code.     

A machine learning approach to query generation in plagiarism source retrieval

Plagiarism source retrieval is the core task of plagiarism detection. It has become the standard for plagiarism detection to use the queries extracted from suspicious documents to retrieve the plagiarism sources. Generating queries from a suspicious document is one of the most important steps in plagiarism source retrieval. Heuristic-based query generation methods are widely used in the current research. Each heuristic-based method has its own advantages, and no one statistically outperforms the others on all suspicious document segments when generating queries for source retrieval. Further improvements on heuristic methods for source retrieval rely mainly on the experience of experts. This leads to difficulties in putting forward new heuristic methods that can overcome the shortcomings of the existing ones. This paper paves the way for a new statistical machine learning approach to select the best queries from the candidates. The statistical machine learning approach to query generation for source retrieval is formulated as a ranking framework. Specifically, it aims to achieve the optimal source retrieval performance for each suspicious document segment. The proposed method exploits learning to rank to generate queries from the candidates. To our knowledge, our work is the first research to apply machine learning methods to resolve the problem of query generation for source retrieval. To solve the essential problem of an absence of training data for learning to rank, the building of training samples for source retrieval is also conducted. We rigorously evaluate various aspects of the proposed method on the publicly available PAN source retrieval corpus. With respect to the established baselines, the experimental results show that applying our proposed query generation method based on machine learning yields statistically significant improvements over baselines in source retrieval effectiveness.     

基于改进蚁群算法UML状态机图的测试用例生成

为了减少测试工作量,防止测试用例数目过多而导致爆炸,提出了使用改进的蚁群算法直接从UML状态机图中生成测试用例的方法.利用UML状态图是对系统可能的生命历程进行建模,由于状态图具有直观性,所以简化了生成测试用例过程,从而减少了测试的工作量.通过构造人工蚁群以及动态更新信息素矩阵,让蚂蚁选择还没有搜索或边信息素最小的顶点,确保了测试用例的数量既满足测试需求,又不会过多.通过编写算法及模拟仿真,对该算法的可行性和有效性做了验证和仿真分析.     

A machine learning method for generation of a neural networkarchitecture: a continuous ID3 algorithm

The relation between the decision trees generated by a machine learning algorithm and the hidden layers of a neural network is described. A continuous ID3 algorithm is proposed that converts decision trees into hidden layers. The algorithm allows self-generation of a feedforward neural network architecture. In addition, it allows interpretation of the knowledge embedded in the generated connections and weights. A fast simulated annealing strategy, known as Cauchy training, is incorporated into the algorithm to escape from local minima. The performance of the algorithm is analyzed on spiral data.     

Live modeling in the context of state machine models and code generation

Software and Systems Modeling - Live modeling has been recognized as an important technique to edit behavioral models while being executed and helps in better understanding the impact of a design...     

Example-based machine translation based on tree–string correspondence and statistical generation

This paper describes an example-based machine translation (EBMT) method based on tree–string correspondence (TSC) and statistical generation. In this method, the translation example is represented as a TSC, which is a triple consisting of a parse tree in the source language, a string in the target language, and the correspondence between the leaf node of the source-language tree and the substring of the target-language string. For an input sentence to be translated, it is first parsed into a tree. Then the TSC forest which best matches the input tree is searched for. Finally the translation is generated using a statistical generation model to combine the target-language strings of the TSCs. The generation model consists of three features: the semantic similarity between the tree in the TSC and the input tree, the translation probability of translating the source word into the target word, and the language-model probability for the target-language string. Based on the above method, we build an English-to-Chinese MT system. Experimental results indicate that the performance of our system is comparable with phrase-based statistical MT systems.