**1. Introduction**

[22] Kara, S., Gu°ven, A., Okandan, M., & Dirgenali, F. Utilization of Artificial Neural Networks and Autoregressive Modeling in Diagnosing Mitral Valve Stenosis. (in

[23] Wright, I. A., Gough, N. A. J., Rakebrandt, F., Wahab, M., & Woodcock, J. P. (1997). Neural Network Analysis of Doppler Ultrasound Blood Flow Signals: A pilot study.

[24] Salvi, M., Dazzi, D., Pelistri, I., Neri, F., & Wall, J. R. (2002). Classification and Predic‐ tion of the Progression of Thyroid-Associated Ophthalmopathy By An Artificial

[25] Nguyen, H. T., Butler, M., Roychoudhry, A., Shannon, A. G., Flack, J., & Mitchell, P. (1996). Classification of Diabetic Retinopathy Using Neural Networks. *18th Annual International Conference of the IEEE Engineering In Medicine And Biology Society, Amster‐*

[26] Tigges, P., Kathmann, N., & Engel, R. R. (1997). Identification of Input Variables for Feature Based Artificial Neural Networks-Saccade Detection in EOG Recordings. *In‐*

[27] Chan, B. C. B., Chan, F. H. Y., Lam, F. K., Lui, P. W., & Poon, P. W. F. (1997). Fast Detection of Venous Air Embolism in Doppler Heart Sound Using the Wavelet

[28] Tu°rkog˘lu, I., Arslan, A., & I˙lkay, E. (2002). An Expert System for Diagnosis of the

[29] Schmitz, G. P. J., Aldrich, C., & Gouws, F. S. (1999). ANN-DT An Algorithm for Ex‐ traction of Decision Trees from artificial Neural Networks. *IEEE Transactions on Neu‐*

[30] Sethi, I. K. (1992). Layered Neural Net Design Through Decision Trees. *Proceedings of*

[31] Quilan, J. (1993). C4.5: Programs for Machine Learning. Morgan Kaufmann, San Ma‐

[32] Fabian, H. P., Chan, K. S., Ho, K. Y., & Leong, S. K. (2004). A Study on Decision Tree.

[33] Haykin, S. (1999). Neural Networks: a Comprehensive Foundation. Second Edition.

[34] Yedjour, D., Yedjour, H., & Benyettou, A. (2011). Explaining Results Of Artificial

[35] Trees, H. V. (2001). Detection, Estimation and Modulation Theory- Part I. John Wiley,

[36] Huseyin, C., & Srinivasan, R. (2004). Configuration of Detection Software: a Compar‐ ison of Decision and Game Theory Approaches. *Decision Analysis*, 1(3), 131-148.

*2nd Engineering & Technology Student's Congress. Kota Kinabalu.: SKTM.*

Neural Networks. *Journel Of Applied Scinces*, 2(3).

Transform. *IEEE Transactions on Biomedical Engineering*, 44(4), 237-245.

Heart Valve Diseases. *Expert Systems with Applications*, 23, 229-236.

press), *Computers in Biology and Medicine*.

*Ultrasound in Medicine and Biology*, 23(5), 683-690.

Neural Network. *Ophthalmology*, 109(9), 1703-8.

*ternational Journal of Medical Informatics*, 45, 175-184.

*dam*, 1548-1549.

84 Advances in Expert Systems

*ral Networks*.

*the IEEE*.

teo, CA.

Prentice Hall., 842.

New York.

Several companies and institutions now realize knowledge as an active relavant for the mar‐ ket organization differentiation. This scenario explains the need for systems that assist the user in the acquisition process and knowledge management. Intelligent systems, known as expert systems (ES) [19] serve to this purpose in the extent that they have signed as facilitators in this process. These are systems that are based on expert knowledge, on any subject, in order to em‐ ulate human expertise in the specific field. To obtain this knowledge, the knowledge engi‐ neers, also called software engineers, need to develop methodologies for intelligent systems. In this area there is still no unified methodology that provides effective methods, notations and tools to aid in development. Among the most used technologies we can mention: KADS [10], MIKE [2] and Protégé [8]. KADS is a structured way of developing these systems that pro‐ vides a special focus on the characteristics and problems of development of the SE. KADS uses the waterfall model as a basis to guide the development and adds refinement stages of use and knowledge [10]. Moreover, the MIKE methodology (Model-Based and Incremental Knowl‐ edge Engineering) makes use of formal specification and semi-formal techniques during the incremental development of the system. The phases of this model are four (Figure 1), being the first, knowledge acquisition, made in a cyclic manner between the subphases of task analysis, model construction and evolution. After the acquisition phase, the design, implementation and evolution are cyclical until the system is built.

© 2012 Boarati et al.; licensee InTech. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. © 2012 Boarati et al.; licensee InTech. This is a paper distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

It is an approach to build domain ontologies and includes the notion of a library of reusable problem-solving methods (PSMs) that perform tasks. In PROTÉGÉ-II, PSM are decomposa‐ ble into subtasks. Other methods, sometimes called sub methods*,* can perform these sub‐ tasks. Primitive methods that cannot be decomposed further are called mechanisms. This decomposition is made to allow the reuse of knowledge, an essential part of the methodolo‐ gy. The choice of which method is suitable for the development of an intelligent system is essential, given the complexity of systems in the field of artificial intelligence (AI). It is im‐ portant to define a process that systematizes the life cycle, allowing a greater skill in eliciting and models description. In this work, it was elaborated a guide for knowledge acquisition based on ontologies [9] and applied to the extension of an expert system of recommenda‐ tions (guidelines) for designing human-computer interfaces.

**2. Knowledge acquisition for the new GuideExpert acquaintance**

the common characteristics and goals [6].

**Figure 2.** GuideExpert expert system

The GuideExpert system is an expert system developed to assist the designer of humancomputer interfaces during the phases of design and evaluation of interfaces. The system consists of five elements: user interface, inference engine, working memory, knowledge base and database. Figure 2 shows the architecture of the expert system. Through a series of questions and screens, the system selects a series of recommendations (called guidelines) of experts in the field of interfaces. The GuideExpert in its first version consisted of three hun‐ dred and twenty six guidelines of elicited interfaces projects from various authors and ex‐ perts in the field of interfaces. To search the knowledge base of the GuideExpert it was defined the meta-guidelines. They are concepts which embrace the guidelines according to

Heuristics for User Interface Design in the Context of Cognitive Styles of Learning and Attention Deficit Disorder

http://dx.doi.org/10.5772/51455

87

In the second phase of the project, it was seen the need to incorporate in the system, the guidelines relative to the diversity of user profiles. We identified several recommendations, heuristic and knowledge about adults, children, handicapped users, users with deficits of at‐ tention and etc. In order to make the knowledge acquisition in this domain it was elaborated a guide based on Ontologies. Ontology is a formal and explicit specification of a shared con‐ ceptualization [9]. The ontologies are used to structure and share the knowledge. They can be seen as the highest level in a hierarchy of knowledge composed of vocabularies, thesauri, taxonomies, ontologies and frames. A taxonomy is to classify information in a hierarchy

The existence of a taxonomy in GuideExpert system, formed by the meta-guidelines, moti‐ vated the ontology conceptualization for projects in human-computer interfaces. For the cre‐

(tree) with the generalization relationship "kind-of" (parent-child) [4].

**Figure 1.** Stages of knowledge acquisition, design, implementation and evolution from MIKE.

This expert system called GuideExpert was expanded to include recommendations about profiles of users with learning disorder (TA), attention deficit disorder / hyperactivity (ADD / H) and advices on cognitive learning styles (ECAS). The learning disorder is defined where individuals can not develop as expected in appropriate age scholl [22], on the other hand the deficit of attention disorder/ hyperactivity and impulsivity [1]. In turn, the cogni‐ tive learning styles represent a categorization of the cognition particularities with their re‐ spective skills [21]. There are several recommendations on how computer interfaces should be designed in order to attend, in a satisfactory way, users with learning disorders and dif‐ ferent cognitive styles, among other features. Thus, the aim of incorporating this knowledge to the GuideExpert base needed the establishment of a process for knowledge acquisition which will be presented in the next section.
