Mining knowledge from natural language texts using fuzzy associated concept mapping

Knowledge Management Research Centre, Department of Industrial and Systems Engineering, The Hong Kong Polytechnic University, Hung Hum, Kowloon, Hong Kong
Information Processing & Management (Impact Factor: 1.27). 09/2008; DOI: 10.1016/j.ipm.2008.05.002
Source: DBLP

ABSTRACT Natural Language Processing (NLP) techniques have been successfully used to automatically extract information from unstructured text through a detailed analysis of their content, often to satisfy particular information needs. In this paper, an automatic concept map construction technique, Fuzzy Association Concept Mapping (FACM), is proposed for the conversion of abstracted short texts into concept maps. The approach consists of a linguistic module and a recommendation module. The linguistic module is a text mining method that does not require the use to have any prior knowledge about using NLP techniques. It incorporates rule-based reasoning (RBR) and case based reasoning (CBR) for anaphoric resolution. It aims at extracting the propositions in text so as to construct a concept map automatically. The recommendation module is arrived at by adopting fuzzy set theories. It is an interactive process which provides suggestions of propositions for further human refinement of the automatically generated concept maps. The suggested propositions are relationships among the concepts which are not explicitly found in the paragraphs. This technique helps to stimulate individual reflection and generate new knowledge. Evaluation was carried out by using the Science Citation Index (SCI) abstract database and CNET News as test data, which are well known databases and the quality of the text is assured. Experimental results show that the automatically generated concept maps conform to the outputs generated manually by domain experts, since the degree of difference between them is proportionally small. The method provides users with the ability to convert scientific and short texts into a structured format which can be easily processed by computer. Moreover, it provides knowledge workers with extra time to re-think their written text and to view their knowledge from another angle.

Download full-text


Available from: W.M. Wang, Jun 22, 2015
14 Reads
  • Source
    • "The problem with linguistic techniques is that they are limited to a specific language, and linguistic resources must exist for used language. The majority of linguistic tools and methods are based on the English language, and researchers mostly use them in CMM [14], [15], [17], [18], [23]–[27], [32]. A number of them use the WordNet for lemmatization, POS tagging and terms disambiguation. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Concept map is a graphical technique for representing knowledge, successfully used in different areas, including education, knowledge management, business and intelligence. In this paper, an overview of different approaches to automatic creation of concept maps from textual and non-textual sources is given. Concept map mining process is defined, and one method for creation of concept maps from unstructured textual sources in the Croatian language is described.
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Acquisition of knowledge must be interwoven with the process of applying it. However, traditional training methods which provide abstract knowledge have shown ineffective for gaining experience of the work. In order to solve this problem, more and more researchers have included narrative in simulation, which is known as narrative simulation. By providing the narratives, participants recognize the choices, decisions, and experience that lead to the consequences of those decisions. It has been proven that narrative simulation is very useful in facilitating in-depth learning and reflective learning. However, conventional methods of data collection and narrative construction for narrative simulation are labor intensive and time consuming. They make use of previous narratives manually and directly. They are inadequate to cope with the fast moving world where knowledge is changing rapidly. In order to provide a way for facilitating the construction of narrative simulation, a novel computational narrative construction method is proposed. By incorporating technologies of knowledge-based system (KBS), computational linguistics, and artificial intelligence (AI), the proposed method provides an efficient and effective way for collecting narratives and automating the construction of narratives. The method converts the unstructured narratives into a structural representation for abstraction and facilitating computing processing. Moreover, it constructs the narratives that combine multiple narratives into a single narrative by applying a forecasting algorithm. The proposed method was successfully implemented in early intervention in mental health care of a social service company in Hong Kong since the case records in that process have structural similarities to narrative. The accuracies of data conversion and predictive function were measured based on recall and precision and encouraging results were obtained. High recall and precision are achieved in the data conversion function, and high recall for the predictive function when new concepts are excluded. The results show that it is possible for converting multiple narratives into a single narrative automatically. Based on the approach, it helps to stimulate knowledge workers to explore new problem solving methods so as to increase the quality of their solutions.
    Expert Systems with Applications 05/2009; 36(4-36):8093-8102. DOI:10.1016/j.eswa.2008.10.035 · 2.24 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Choosing the optimal terms to represent a search engine query is not trivial, and may involve an iterative process such as relevance feedback, repeated unaided attempts by the user or the automatic suggestion of additional terms, which the user may select or reject. This is particularly true of a multimedia search engine which searches on concepts as well as user-input terms, since the user is unlikely to be familiar with all the system-known concepts. We propose three concept suggestion strategies: suggestion by normalised textual matching, by semantic similarity, and by the use of a similarity matrix. We have evaluated these three strategies by comparing machine suggestions with the suggestions produced by professional annotators, using the measures of micro- and macro- precision and recall. The semantic similarity strategy outperformed the use of a similarity matrix at a range of thresholds. Normalised textual matching, which is the simplest strategy, performed almost as well as the semantic similarity one on recall-based measures, and even better on precision-based and F-based measures.
    Computer Science and Information Technology, 2009. IMCSIT '09. International Multiconference on; 11/2009
Show more