Visualising Fuzzy Cognitive Maps Yuan Miao School of Engineering and Science Victoria University, Melbourne, Australia Yuan.Miao@vu.edu.au Abstract— Visualised presentation of cognitive model can greatly promote the easy comprehension of the model, leading to the wide application of Fuzzy Cognitive Maps (FCM). This paper improves the visualisation of FCM from normal weighted digraph to graphs that can differentiate the strength of causal linkage and the strength of the concepts. The advantage of the new model is apparently illustrated through comparing the new model with the existing ones for the FCMs in the recent literature. The visualised FCM is also applied in word cloud analysis on text term frequency presentation, which improves the knowledge modeling through presenting the key linkage among text terms. Keywords- Fuzzy Cognitive Map, Word Cloud, Visualisation I. INTRODUCTION In a wide range of information application domains, a major goal of machine intelligence or human machine hybrid intelligence is to capture human knowledge. If human beings can articulate their cognition, it is then possible for computing models to capture this cognition and facilitate communication, collaboration and automated services. Fuzzy Cognitive Map (FCM) is a tool to visually represent human cognitions [2]. The following figure is an example FCM fragment. Figure 1.1 FCM of Interest Rate and Investment The FCM fragment in Figure 1.1 represents the following cognition: - interest rate has negative impact on the investment expansion; a high interest rate discourages investment and a low interest rate encourages investment; - investment sentiment represents the willingness or intention to invest, which normally came from actual profits; investment sentiment has positive impact on the investment expansion; a high investment sentiment encourages investment; - the causal link from interest rate to investment expansion is very strong; and - the causal link from investment sentiment to investment expansion is mild. As compared with other more formal modeling tools, FCM is easy to use and graphically comprehensible. It has attracted many domain experts to represent their cognition or cognitive knowledge. The applications span over a wide range of areas, such like software quality risk analysis[3], drought management[4], analysing dynamics in logistics[5], combating diabetes [12], anti-terrorism [7], energy consumption [9], ecosystems [8][10], medical decision support [14], and games based learning [13]. It has been discovered that a transformation [11] exists among a number of major FCM variations [6], indicating that different presentation forms are important to FCM applications. The following are three recent example applications of FCM. 1) Giles and et al [12] have used FCM to integrate conventional science and aboriginal perspectives to study the development of type 2 diabetes. Figure 1.2 is a FCM used in the study, which shows the causal links among diabetes related factors like body weight, employment and diets. The FCM presentation of domain experts cognition become a basis for their analysis and study on the development of type 2 diabetes. - Figure 1.2 FCM on contributing factors to type 2 diabetes [12] 2) Luo and et al [13] used FCM to model factors in game based teaching and learning. Figure 1.3 is a FCM used in their study in modeling high level robot planning. Similar FCMs are used in their game based driver training system. This work is supported by Australia Research Council (LP100100624) and Singapore National Research Foundation (NRF IDM2- 12). U.S. Government work not protected by U.S. copyright WCCI 2012 IEEE World Congress on Computational Intelligence June, 10-15, 2012 - Brisbane, Australia FUZZ IEEE