Recently, designers have been required solving comprehensive problem becoming greater and more complicated. In relation to this background, we have proposed Universal Abduction Studio (UAS); a computer environment that synthetically supports creative design. However, it is difficult for a designer to manually acquire multiple domain knowledge required for UAS. Therefore, we propose a Web-based knowledge database construction method for supporting design by UAS in this paper.
To accomplish knowledge intensive tasks, people in organizations must be able to find the knowledge or information needed to solve complex problems. For this, people often rely on their past experiences, explicit documents, and others who have the needed expertise. Knowledge Management Systems that enhance and facilitate the process of finding the right expert in an organization have gained much attention in recent years. This paper explores the potential benefits and challenges of using ontologies for improving existing systems. A modeling technique from requirements engineering is used to evaluate the proposed system and analyze the impact it would have on the goals of the stakeholders. This paper also discusses the organizational settings required for the successful deployment of the system in practice.
The primary purpose of the paper is to examine the ways in which social information affects knowledge-sharing behavior in an organization. Based on theory of reasoned action and focusing on knowledge sharing setting, we propose that subjective norms and attitudes influence behavioral intention. Three processes drawing from social information processing theory (i.e., internalization, identification, and compliance) are postulated as antecedents of the intention to share knowledge. We also posit that knowledge type (as a moderator) intervenes the forming patterns of sharing behavior. Structural equation modeling was used to test hypotheses. Empirical data are collected from 229 respondents and our arguments were statistically supported. Some theoretical and practical implications are also discussed.
This paper investigated the correlation between the level of knowledge management (KM) and efficiency of ceramic tile companies in Iran. KM was measured with four components: infrastructure, process, people, and strategy and then efficiency was measured with data envelopment analysis (DEA). First, factor analysis was used in order to determine KM factors. Then, the partial correlation between KM components and efficiency was computed. Finally, multiple linear regressions were computed with the four independent variables (factors) and efficiency. Results showed positive relationship between the KM level and the efficiency in ceramic tile industry.
Knowledge represents the most important production factor for huge fields of today's economy. Since more than a decade the research on knowledge management and intellectual capital management has brought insights for the question of how organizations and their employees create value and eventually also profits. As human capital is highly volatile, organizations strive to transfer employee's knowledge into the more institutionalized structural capital. For this transformation as well as to identify and fill knowledge gaps we propose the integration of the ICRB framework and the eduWEAVER approach to a comprehensive IT-based management approach for knowledge management.
This paper proposes a method to extract causal knowledge (cause and effect relations) using clue phrases and syntactic patterns from Japanese newspaper articles concerning economic trends. For example, a sentence fragment "World economy recession due to the subprime loan crisis ..." contains causal knowledge in which "World economy recession" is an effect phrase and "the subprime loan crisis" is its cause phrase. These relations are found by clue phrases, such as "ため(tame: because)" and "により(niyori: due to)". We, first, investigated newspaper corpus by annotating causal knowledge and clue phrases. We found that some specific syntactic patterns are useful to improve accuracy to extract causal knowledge. Finally, we developed our system using the clue phrases and the syntactic patterns and showed the evaluation results on a large corpus.
Users in ubiquitous environments can use dynamic services whenever and wherever they are located because these environments connect objects and users through wire and wireless networks. Also, there are many devices and services in these environments. However, it is difficult to effectively use conventional filtering method of the recommendation system in future ubiquitous environments because it does not reflect context information well in these environments. This paper attempt to define context model and propose new Collaborative Filtering (CF) based on Hidden Markov Models (HMMs) that are trained by context information. The Collaborative Filtering using HMMs (CFH) is suited to a user's interests and preferences. The Ubiquitous Recommendation System (URS) used in this study based on CFH uses an Open Service Gateway Initiative (OSGi) framework to recognize context information and connect device in smart home.
The paper starts with analysing the requirements of supporting largely unstructured, collaboration-oriented processes as opposed to highly structured, workflow-based processes. The approach derived from those requirements utilizes the concept of interaction patterns to impose some control and guidance on a collaboration process without reducing the needed flexibility. This not only causes a much greater process transparency but also allows to provide and capture knowledge within the collaboration interactions. The application-specific instantiation of the interaction patterns and their enrichment with integrity constraints is done using a model-driven approach. The resulting application model is also a knowledge artefact which represents an important fragment of domain knowledge.
In collocated software development teams, informal communication is the key enabler for sharing knowledge. In distributed teams, development infrastructures have to fill communication gaps with light-weight articulation and sharing facilities for evolving development knowledge. We propose an ontology-based framework to capture, access and share developers' experiences in a decentralized, contextualized manner. Capturing developers' interaction with related artifacts and providing a Wiki-like annotation approach triggers knowledge capture. Integrated semantic search and recommendation fosters knowledge access and sharing. Our framework enables distributed teams to become more effective by learning from each other's experiences, e.g. on reusing specific components and handling semantic errors.
This paper recommends a structure to represent and a method to retrieve knowledge artifacts for repository-based knowledge management systems. We describe the representational structure and explain how it can be adopted. The structure includes a temporal dimension, which encourages knowledge sharing during knowledge creation. The retrieval method we present is designed to benefit from the representational structure and provides guidance to users on how many terms to enter when creating a query to search for knowledge artifacts. The combination of the structure and the retrieval method produces an adequate strategy for knowledge sharing that guides targeted users toward best results.
With a rich variety of forms and types, digital resources are complex data objects. They grows fast in volume on the Web, but hard to be classified efficiently. The paper presents a practical classification solution using features from file names and extensions of digital resources. The features are easy to get and common to all resource. But they are generally low frequency and sparse, which implies that statistical approach may not work well. Our solution combines Naive Bayes (NB) classifier with Simple Good-Turing (SGT) probability estimation, which shows great promise for this condition with a total accuracy of 80%. In our opinion, the results are due to 1) the features fit the NB's conditional independence hypothesis well; 2) the abound one-time-occurrence features lead to reasonable probability estimation on unobserved features, which also means general feature selection strategy is not needed in this case. A 7.4TB digital resource collection, CDAL, is used to train and evaluate the model.
In this paper, we propose to improve the overall process of information retrieval by explicitly addressing information provision from private spaces of individual users into the public information space of an organization. Therefore, we present our approach of inverse search , which aims to stimulate the diffusion of documents from these private spaces. We introduce the notion of an organizational information need (OIN) based on query logs and further usage statistics of our system. This information is used to recommend people to share private documents containing relevant information. Our main contributions are describing means to identify documents that should be shared and a framework to foster the diffusion of such documents. We also describe the implementation and results from initial evaluation studies.
In recent automated and integrated manufacturing, so-called intelligence skill is becoming more and more important and its efficient transfer to the next generation is a key to keeping a factory productive and competitive. But, currently, it needs costly on-the-job training (OJT) and it is crucial to reduce its cost. In this paper, we propose a new approach without OJT, that is, combinational usage of ontologies and a rule-based system. It helps domain experts externalize their tacit intelligence skill and helps novices internalize it.
To support data mining post-processing, which is one of the important procedures in a data mining process, at least 40 indices are proposed to acquire valuable knowledge. However, since their behaviors have never been elucidated, domain experts are required to spend their time to understanding the meanings of each index in a given data mining result. In this paper, we present an analysis of the behavior of objective rule evaluation indices on classification rule sets by principle component analysis (PCA). Therefore, we carried out a PCA to a dataset consisting of the 39 objective rule evaluation indices. In order to obtain the dataset, we calculated the average values of the bootstrap method on 32 classification rule sets learned by information gain ratio. Then, we identified the seven functional groups of the objective indices based on the PCA. Using this result, we discuss a rule evaluation interface for use by human experts.
In this paper, we will present some prospective ideas which should allow firms' strategic positioning of the market by using knowledge as a key strategic leverage. After presenting the three basic theories underlined design for learning and teaching, paper continues by describing the basic model based upon which we will develop our unique model; knowledge ingenition process. A generic framework is proposed containing a macro-model and a set of micro-models mapping knowledge elements and their dependencies. These models together are necessary to analyze the knowledge situation of the firm and to conceive a roadmap for future trainings of various employees of the firm during the lifecycle of a product. These concepts are illustrated through a part of a case study.
Technical literature such as patents, research papers, whitepapers, and technology news articles are widely recognized as important information sources for people seeking broad knowledge in technology fields. However, it is generally a labor intensive task to survey these resources to track major advances in a broad range of technical areas. To alleviate this problem, we propose a novel survey assistance tool that focuses on a novel semantic class for phrases, advantage phrases, which mention strong, advantageous points of technologies or products. The advantage phrases such as "reduce cost," "improve PC performance," and "provide early warning of a future failure" can help users to grasp the capabilities of a new technology and to come up with innovative solutions with large business values for themselves and their clients. The proposed tool automatically extracts and lists up those advantage phrases from large technical documents, and places the phrases that mention novel technology applications high on the output list. The developed prototype of the tool is now available for consultants analyzing patent disclosures. In this paper, a method to identify advantage phrases in technical documents and a scoring function to give a higher score to novel applications of a technology are proposed and evaluated.
The scientific breakthroughs resulting from the collaborations between researchers often outperform the expectations. But finding the partners who will bring this synergic effect can take time and sometime gets nowhere considering the huge amounts of experts in various disciplines. We propose to build a link predictor in a network where nodes represent researchers and links - coauthorships. In this method we use the structure of the constructed graph, and propose to add a semantic and event based approach to improve the accuracy of the predictor. In this case, predictors might offer good suggestions for future collaborations. We will be able to compute the classification of a massive dataset in a reasonable time by under-sampling and balancing the data. This model could be extended in other fields where the research of partnership is important as in world of institutions, associations or companies. We believe that it could also help with finding communities of topics, since link predictors contain implicit information about the semantic relation between researchers.
In this talk, various trends in the current service economy are explained. Global economy has turned out to be a service-oriented economy, in addition to the fact that the economy is going along with more digitized information rather than just labor intensive human service activities. In Japan's case, for example, more than 70% of GDP has been created by service sectors in a broader sense recently, and the ratio is still going up. However, the problem we are facing to the service-oriented economy emerges due to the uncertainness of service management and the low productivity of service businesses in general. We explain how knowledge exploratory could contribute to providing a solution or an insight to improve the situation as "service innovation", and to discuss framework and/or design process of service innovation, illustrating with some of the actual research and education activities. They include service literacy management, service blueprinting based on knowledge management framework and common data models such as UML. The goal of this talk is to provide a common view of the knowledge management framework that will support service innovation for the service-oriented economy. We believe that this kind of systematic approach, having human beings located in the system, will have more meaningful implications for the new economy.
In this paper, we consider long documents and try to find differences between document collections. In the analysis of document collections such as project status reports or annual reports, each document and each sentence tend to be relatively long. Therefore, it can be difficult to derive insights by looking only for representative concepts in the selected document collection based on a divergence metric. In this paper, we propose an analysis approach based on contextual information. By extracting pairs of a topic word and a keyword and assessing their representativeness in the selected document collection, we are developing a method to extract insights from these long documents. Applying the proposed method for the analysis between the annual reports of bankrupt companies and those of sound companies, we were able to derive insights that could not be extracted with the conventional methods.
