Raymondus Kosala

Web mining research: a survey

With the huge amount of information available online, the World Wide Web is a fertile area for data mining research. The Web mining research is at the cross road of research from several research communities, such as database, information retrieval, and within AI, especially the sub-areas of machine learning and natural language processing. However, there is a lot of confusions when comparing research efforts from different point of views. In this paper, we survey the research in the area of Web mining, point out some confusions regarded the usage of the term Web mining and suggest three Web mining categories. Then we situate some of the research with respect to these three categories. We also explore the connection between the Web mining categories and the related agent paradigm. For the survey, we focus on representation issues, on the process, on the learning algorithm, and on the application of the recent works as the criteria. We conclude the paper with some research issues.

Information Extraction in Structured Documents Using Tree Automata Induction

Information extraction (IE) addresses the problem of extracting specific information from a collection of documents. Much of the previous work for IE from structured documents formatted in HTML or XML uses techniques for IE from strings, such as grammar and automata induction. However, such documents have a tree structure. Hence it is natural to investigate methods that are able to recognise and exploit this tree structure. We do this by exploring the use of tree automata for IE in structured documents. Experimental results on benchmark data sets show that our approach compares favorably with previous approaches.

Information extraction from structured documents using k -testable tree automaton inference

Information extraction (IE) addresses the problem of extracting specific information from a collection of documents. Much of the previous work on IE from structured documents, such as HTML or XML, uses learning techniques that are based on strings, such as finite automata induction. These methods do not exploit the tree structure of the documents. A natural way to do this is to induce tree automata, which are like finite state automata but parse trees instead of strings. In this work, we explore induction of k-testable ranked tree automata from a small set of annotated examples. We describe three variants which differ in the way they generalize the inferred automaton. Experimental results on a set of benchmark data sets show that our approach compares favorably to string-based approaches. However, the quality of the extraction is still suboptimal.

Color quick response code for mobile content distribution

In this paper, we propose Color Quick Response Code or CQR Code for short, which is a Matrix code that has been extended into a 3D barcode system. Matrix code itself is a 2-dimensional barcode that appears to be a 2D image that consists of usually black and white dots that can be printed or displayed on a screen. CQR Code is an extended version of the popular QR Code. 3D barcodes contains data not only in its x and y axis, but also in depth. By stacking QR Codes we can create a new barcode system that can store more data than other existing barcode systems. Through this barcode system, we can embed data that normally could not be done using existing barcode systems. From the tests that had been conducted, currently the CQR Code is capable of storing data up to 9KB with 3 layers and 8 colors. The CQR Code can be read with 40% - 100% chance of success using a multi-format mobile phone barcode reader prototype. We have also implemented a simple color correction algorithm into the barcode reader to ensure its accuracy. In conclusion, CQR Code has the potential for future developments for it to be a simple, sophisticated, and easy to use barcode system that can store relatively large amounts of data.

Comparing the effectiveness of digital contents for improving learning outcomes in computer programming for autodidact students

Digital content makes improvement in e-learning for transferring knowledge to learners. By e-learning that delivers digital contents, learners can learn anywhere and anytime and educators can deliver materials in an interactive way. Digital content becomes very important in improving self-learner understanding in computer programming. Experimental design method using control-group pretest-posttest is conducted to compare digital contents effectiveness of learning outcomes in computer programming field for self-learning students. The instrument used is a programming case, with factor and level to be tested is various digital content types, and desired response is learning outcomes increment. This study concludes that there is an average increment of learning outcomes by using various digital contents for self-learning, but still lower average increment compared with traditional classroom method. Results of this study can be a recommendation for educators and students in applying and determining the use of digital contents for their self-learning.

Knowledge-based integrated financial forecasting system

This paper seeks to implement and test a financial forecasting agent which employs time series, derived time series data, and news that are retrieved and extracted from the Web. This research focuses on the time series data of some individual stocks from the Indonesian Stock Exchange as well as the index data. The financial forecasting agent implemented is based on a Multilayer Neural Network trained with the Genetic Algorithm. An incrementally trained Naïve Bayesian classifier to classify the news stories automatically is also implemented. The agent is measured for its performance on the specified historical stock price data. The first testing surveys its performance against the benchmark agents in a range of stocks. Although the agent does not consistently outperform the benchmarks, it does have its own advantages in terms of stability. The second test tries to find out whether news improves the agents prediction accuracy. We find that in certain stocks and the index, the inclusion of news does improve the prediction accuracy.

A Comparative Study of Skeletal and Keyframe Animations in a Multiplayer Online Game

As the game development process is a very tedious one, a good knowledge of which game animation technique that work best in practice is very essential. In this paper we study the performance of two popular animation models – the skeletal and key frame animation models. We have done some experiments using a test bed program to measure the performance, memory usage and hard disk space used by the two animation models. Our experiment results suggest that the skeletal model has major advantages over the key frame model, both in terms of the FPS (frame per second) and the memory load. We conclude that the skeleton or bone based animation models are more suitable for multiplayer online role playing games (MORPG), where the bandwidth usage is very intensive.

Software size measurement of knowledge management portal with use case point

Knowledge Management portal is a system to support Knowledge Management process, in order to create, capture, develop, share, reuse and optimize the knowledge and particularly in Bina Nusantara University which has implemented Knowledge Management System (KMS) since 2002. However, this KMS need to be measured in order to know how better this KMS in term of the software size. The BINUS KMS will be measured in term of their software size in functionality perspective with use case point method. This metric of KMS will be used by management to know how better the software size, complexity level and effort to development in numbering. Measurement of software size with software metric such as Use Case Point upon use case diagram for BINUS knowledge Management Portal shows that the project has medium software size with score Use Case Point (UCP) = 108.56 and has estimate effort will be developed in 2,064 hours (or in 258 days or 51.6 weeks or 12.9 months) and has development cost for 516,000,000.00 rupiah (Indonesian currency). Use Case Point, estimate effort and project value will powerful to help management in order to make decision regarding the implementation of IT software project development in term of time, money and people.

Two-Way ANOVA with interaction approach to compare content creation speed performance in knowledge management system

Speed performance is important in the process of creating content on the Knowledge Management System. Bina Nusantara (BINUS) University has implemented Knowledge Management System (KMS) since 2002. This study examined the speed performance of content creation in four modules of BINUS University KMS: Documents, Video, Material, and Binuspedia. The examined to get mean value of speed performance was conducted in 3 location campuses of BINUS University: Anggrek Campus, Syahdan Campus and Alam Sutera Campus, each was done with five repetition. The speed performance was analyzed using Two-way ANOVA with interaction approach. According to the experiment results, F calculated for module is greater than F table, F calculated for campus is less than F table and F calculated for interaction is less than F table. Thus, the conclusions are: (1) there are differences in the average value of speed performance for each module, (2) there is no difference in the average value of speed performance for each campus, and (3) there is no interaction between campuses and modules on the speed performance.