Catálogo de publicaciones - libros

How well the privacy policy follows a regulation is one of the current concerns of the user. Such a task can be accomplished by directly querying the policy statement with the regulation text. Automation of the process requires an expressive meaning-based framework for Natural Language Processing (NLP). This paper discusses the Ontological Semantics approach to the issue of verifying compliance and illustrates the potential of utilizing the framework in the domain of Privacy management for NLP-related tasks. As an example a section from BCBS and corresponding HIPAA regulations are used.

This study is mainly to evaluate measurement models for web purchasing intention, which reflects to a tendency of attitudes and behaviors toward the online purchasing behaviors. Four dimensions of web purchasing intention are proposed based on a literature review, including (1) Information Provision: product-related information that e-retailing users receive from the Internet; (2) Alternative Evaluation: issues that affect an individual’s decision making when purchasing online; (3) Customer Service: services that e-retailing websites offer for customers in the e-transactions and post-sale process; (4) Price: the final dealing prices of products on e-retailing websites. This study developed a questionnaire and delivered to consumers with web purchasing experience. The method of Structural Equation Modeling (SEM) is adopted to verify the internal quality of the proposed measurement.

This paper describes web application for data analysis by means of evidential dominance. Red Hat Linux 9, Apache, Tomcat, Java 2 Platform, and Eclipse are used to this system as tools. In conventional system, Java for Evidential Dominance is applied as Java servlet. Client accesses to the Web Server to analyse data. Java for evidential dominance is started and implemented on Web Server. The server processes and responses the result of analysis to the client. The information of data is not remained in server. This system not only responses the result of analysis to the client, but also save the information of data in server. Evidential dominance is applied to decide the priority relation of data which varies widely and have little amount of data. Evidential dominance not only is effective to focus on one phase a lot of alternatives, but also applies to decision making problem.

There is almost universal agreement on the benefits of an Electronic Medical Records (EMR) System, yet successful implementations are few. While EMR systems focus on the important areas of functionality, interoperability and security, the area of usability has been overlooked. No electronic medical records system can be viable unless it addresses usability. We have had the opportunity to work with several systems and have employed field data gathering, expert reviews, usability testing and ScoreCarding to enhance the usability of EMR systems. Unless usability becomes more widespread and is given a higher priority, the benefits of an EMR system will remain elusive.

Usability testing on Ilmu (i.e. WebOPAC for children) showed that poor graphic design contributed the most to the usability problem. Three main issues of graphic design are layout, use of colors and human visual perception. A design approach is formulated in an effort to develop a conceptual design that specifies children’s requirements and preferences on graphic issues. This involves two main activities which are identifying children’s preferences on graphic design and generating design ideas using participatory design method. Four categories of requirements and preferences on graphic design are derived from children. Each category is demonstrated by design ideas. All these constitute towards the formulation of a conceptual design for children’s WebOPAC interface and become the basis for redesigning Ilmu’s interface.

This study evaluated users’ behaviors when performing inexpensive or expensive e-commerce purchases on familiar and unfamiliar Web sites. Users were more comfortable with making inexpensive than expensive purchases. They also felt more secure and that their privacy was better protected when shopping with a familiar Web site than an unfamiliar one, especially for expensive purchases. For inexpensive purchases, if the price was “right”, participants were willing to purchase the product on unfamiliar Web sites. For expensive purchases, though, the reputation of the organization hosting the Web site was the most important factor. In both cases, privacy was a minor determinant for deciding whether to make a purchase from a Web site. Only 20% of the users regularly accessed the sites’ privacy policies during their interactions with the sites. Moreover, less than half of the participants even looked at privacy-policy links during their interactions with the Web sites.

This study examined participants’ comprehension of Web privacy policies when the information from the policy had to be recalled from memory or when participants were able to view the policy while searching for the answers to specific questions. Eye-gaze data were analyzed to examine where users focus their attention when reading privacy policies and searching for information in them. Overall, participants showed poor comprehension of the information conveyed in the privacy policies even though they were written at the participants’ level of education. When searching for information in the privacy policies, participants relied on the listing of individual sections provided at the beginning of a privacy policy, when available. When the listing of sections was not available, participants skimmed the entire policy, examining the headings and first few words of each paragraph to determine whether the sections or paragraphs were likely to contain the information for which they were searching.

When using the bookmark or “Favorites” function included with most web browsers, it is difficult to retrieve a target bookmark if it is not properly categorized or tagged with meaningful keywords. For this project, we propose using a method of human-like long term memory (LTM) rather than directories or keywords for searching and managing bookmark data. This experimental model takes into account the user’s browsing experience (i.e. the bookmark’s added date, number of visits, and last date visited) to create and manage bookmarks. A pilot study of the model shows that the new system is significantly more efficient at retrieving information from large collections of bookmark data.

This paper recognizes the need for new approaches in IT based education. Our proposed model depends on two major concepts. The first is to believe in the learning approach not the teaching approach that is widely used. Learning delivery requires providers to alter their in-class approach and demands students to be more independent learners. The second concept is the interdisciplinary learning approach rather than the silo teaching model used currently. For Example, software engineers will not be allowed to work in isolation from the business people, the users, the client, etc. in the real world. Therefore, their college education must qualify them, as much as possible, to deal with the real world. This paper proposes an interdisciplinary project based learning model focusing on collaborative learning (CL) and Business Integration (BI) using the SEE [1] portal. This integral solution is designed to overcome the teaching and silo models used currently.

Today’s businesses increasingly rely on vast amount of information. Yet effective use of information is becoming more and more difficult. This paper describes a general purpose analytics solution, , which embeds two major classes of information analytics techniques and a unique set of visualizations to mine the available information and uncover critical business insights and enhance business performance. The two major classes of analytics technologies include the “taxonomy” analysis and the “relationship” analysis to facilitate understanding and detection of hidden concepts and patterns buried in the information respectively. The BIW technologies have been successfully applied in many application domains, e.g., Customer Relationship Management (CRM) for customer satisfaction analysis, Intellectual Property (IP) for patent portfolio analysis and licensee identification, and Healthcare Life-sciences (HCLS) for facilitating drug discovery by identifying the relationships among chemicals, DNA, proteins, drugs, and diseases. We show some BIW sample applications in this paper.