ICT-Article
Permanent URI for this collectionhttps://repository.li.mahidol.ac.th/handle/20.500.14594/55
Browse
Recent Submissions
Publication Open Access Large scale detailed mapping of dengue vector breeding sites using street view images(2019) Haddawy, Peter; Poom Wettayakorn; Boonpakorn Nonthaleerak; Myat Su Yin; Anuwat Wiratsudakul; Johannes Scho¨ning; Yongjua Laosiritaworn; Balla, Klestia; Sirinut Euaungkanakul; Papichaya Quengdaeng; Kittipop Choknitipakin; Siripong Traivijitkhun; Erawan, Benyarut; Thansuda Kraisang; Reiner, Robert C.; Mahidol University. Faculty of Information Communication Technology; University of Bremen, Germany. Bremen Spatial Cognition Center; Mahidol University. Faculty of Veterinary Science; Ministry of Public Health; University of Camerino. Computer Science Department, School of Science and TechnologyTargeted environmental and ecosystem management remain crucial in control of dengue. However, providing detailed environmental information on a large scale to effectively target dengue control efforts remains a challenge. An important piece of such information is the extent of the presence of potential dengue vector breeding sites, which consist primarily of open containers such as ceramic jars, buckets, old tires, and flowerpots. In this paper we present the design and implementation of a pipeline to detect outdoor open containers which constitute potential dengue vector breeding sites from geotagged images and to create highly detailed container density maps at unprecedented scale. We implement the approach using Google Street View images which have the advantage of broad coverage and of often being two to three years old which allows correlation analyses of container counts against historical data from manual surveys. Containers comprising eight of the most common breeding sites are detected in the images using convolutional neural network transfer learning. Over a test set of images the object recognition algorithm has an accuracy of 0.91 in terms of F-score. Container density counts are generated and displayed on a decision support dashboard. Analyses of the approach are carried out over three provinces in Thailand. The container counts obtained agree well with container counts from available manual surveys. Multi-variate linear regression relating densities of the eight container types to larval survey data shows good prediction of larval index values with an R-squared of 0.674. To delineate conditions under which the container density counts are indicative of larval counts, a number of factors affecting correlation with larval survey data are analyzed. We conclude that creation of container density maps from geotagged images is a promising approach to providing detailed risk maps at large scale.Publication Metadata only Placing pipeline stages on a grid: single path and multipath pipeline execution(2010-01) Ekasit Kijsipongse; Sudsanguan Ngamsuriyaroj; สุดสงวน งามสุริยโรจน์; Mahidol University. Faculty of Information and Communication TechnologyIn a Grid computing environment, several applications such as scientific data analysis and visualization are naturally computation and communication intensive. These applications can be decomposed into a sequence of pipeline stages which can be placed on different Grid nodes for concurrent execution. Due to the aggregation of the computation and communication costs involved, finding the way to place such pipeline stages on a Grid in order to achieve the maximum application throughput becomes a challenging problem. This paper proposes a solution that considers both the pipeline placement and the data movement between stages. Specifically, we try to minimize the computation cost of the pipeline stages while preventing the communication overhead between successive stages from dominating the entire processing time. Our proposed solution consists of two novel methods. The first method is single path pipeline execution, which exploits only temporal parallelism, and the second method is multipath pipeline execution, which considers both temporal and spatial parallelism inherent in any pipeline applications. We evaluate our work in a simulated environment and also conduct a set of experiments in a real Grid computing system. When compared with the results from several traditional placement methods, our proposed methods give the highest throughputPublication Metadata only Inference of lexical ontologies. The LeOnI methodology(2010-01) Farreres, Javier; Gibert, Karina; Rodríguez, Horacio; Charnyote Pluempitiwiriyawej; ชาญยศ ปลื้มปิติวิริยะเวช; Mahidol University. Faculty of Information and Communication Technology. Department of Computer Science.In this article we present a method for semi-automatically deriving lexico-conceptual ontologies in other languages, given a lexico-conceptual ontology for one language and bilingual mapping resources. Our method uses a logistic regression model to combine mappings proposed by a set of classifiers (up to 17 in our implementation). The method is formally described and evaluated by means of two implementations for semiautomatically building Spanish and Thai WordNets using Princeton’s WordNet for English and conventional English–Spanish and English–Thai bilingual dictionaries.Publication Metadata only ระบบอ่านข้อมูลบนแบบฟอร์มภาษีเงินได้บุคคลธรรมดา ภ.ง.ด.91(2009-01) Sukanya Phongsuphap; สังวาล อินต๊ะสุข; Sukanya Phongsuphap; Sangwan Intasuk; สุกัญญา พงษ์สุภาพ; มหาวิทยาลัยมหิดล. คณะเทคโนโลยีสารสนเทศและการสื่อสารงานวิจัยนี้เสนอระบบสำหรับอ่านข้อมูลตัวเลขบนภาพแบบฟอร์มภาษีเงินได้บุคคลธรรมดา ภ.ง.ด.91 โดยพิจารณาทั้งตัวเลขที่เป็นตัวพิมพ์และลายมือเขียน การทำงานของระบบแบ่งออกเป็น 6 ขั้นตอน ดังนี้ 1) การระบุประเภทของแบบฟอร์ม 2) การแยกบล็อกองค์ประกอบของแบบฟอร์ม 3) การแยกแยะบล็อกของฟิลด์ข้อมูลตัวเลข 4) การแยกส่วนของตัวเลข 5) การหาลักษณะสำคัญ ของตัวเลข และ 6) การรู้จำตัวเลข จากการทดลองให้ระบบอ่านข้อมูลจากภาพแบบฟอร์มภาษี จำนวน 100 ภาพ ประกอบด้วยแบบฟอร์มภาษีเงินได้บุคคลธรรมดา ภ.ง.ด.91 จำนวน 50 ภาพ และแบบ ฟอร์มภาษีประเภทอื่น ๆ จำนวน 50 ภาพ ระบบสามารถระบุประเภทของแบบฟอร์มภาษีได้ถูกต้อง ทั้งหมด สามารถแยกบล็อกองค์ประกอบของแบบฟอร์มได้ถูกต้องทั้งหมด ส่วนการแยกส่วน ของตัวเลขสามารถตัดแยกได้ถูกต้องเฉลี่ย 98.06 เปอร์เซ็นต์ (ถูกต้อง 8,193 ตัว จากจำนวนทั้งสิ้น 8,355 ตัว) และสามารถรู้จำตัวเลขได้ถูกต้องเฉลี่ย 95.09 เปอร์เซ็นต์ (ถูกต้อง 7,945 ตัว จากจำนวน ทั้งสิ้น 8,355 ตัว)Publication Metadata only Uncertain inference control in privacy protection(2009) Xiangdong. An; Dawn. Jutla; Nick. Cercone; Charnyote Pluempitiwiriyawej; Hai. Wang; ชาญยศ ปลื้มปิติวิริยะเวช; Mahidol University. Faculty of Information and Communication TechnologyContext management is the key enabler for emerging context-aware applications, and it includes context acquisition, understanding and exchanging. Context exchanging should be made privacy-conscious. We can specify privacy preferences to limit the disclosure of sensitive contexts, but the sensitive contexts could still be derived from those insensitive. To date, there have been very few inference control mechanisms for context management, especially when the environments are uncertain. In this paper, we present an inference control method for private context protection in uncertain environments.Publication Metadata only Thai herb leaf image recognition system (THLIRS)(2011) Chomtip Pornpanomchai; Supolgaj Rimdusit; Piyawan Tanasap; Chutpong Chaiyod; ชมทิพ พรพนมชัย; Mahidol University. Faculty of Information and Communication Technology.There are many kinds of Thai herb species, so it very difficult to identify them all. The objective of this research was to build a computer system that could recognize some Thai herb leaves, using a process called the Thai herb leaf image recognition system (THLIRS). The system consisted of four main components: 1) image acquisition, 2) image preprocessing, 3) recognition and 4) display of results. In the image acquisition component, the system used a digital camera to take a leaf picture with white paper as the background. A one-baht coin was photographed beside the leaf in order to provide a scale for comparison. In the image preprocessing component, the system applied several image-processing techniques to prepare a suitable image for the recognition process. In the recognition component, the system extracted 13 features from the leaf image and used a k-nearest neighbor (k-NN) algorithm in the recognition process. In the result display component, the system displayed the results of the classification. The experiment involved 32 species of Thai herbs, with more than 1,000 leaf images. The system was trained with 656 herb leaf images and was tested using 328 leaf images for a training dataset and 30 leaf images for an untrained dataset. The precision rate of the THLIRS of the training dataset was 93.29, 5.18 and 1.53% for match, mismatch and unknown, respectively. Moreover, the precision rate of the THLIRS of the untrained data set was 0, 23.33 and 76.67% for match, mismatch and unknown, respectively.Publication Metadata only Gaussian kernel approximation algorithm for feedforward neural network design(2009-12-01) Srisuphab, A.; Mitrpanont, J.L.; อนันต์ ศรีสุภาพ; เจริญศรี มิตรภานนท์; อนันต์ ศรีสุภาพ; เจริญศรี มิตรภานนท์; Mahidol University. Faculty of Information and Communication TechnologyA Gaussian kernel approximation algorithm for a feedforward neural network is presented. The approach used by the algorithm, which is based on a constructive learning algorithm, is to create the hidden units directly so that automatic design of the architecture of neural networks can be carried out. The algorithm is defined using the linear summation of input patterns and their randomized input weights. Hidden-layer nodes are defined so as to partition the input space into homogeneous regions, where each region contains patterns belonging to the same class. The largest region is used to define the center of the corresponding Gaussian hidden nodes. The algorithm is tested on three benchmark data sets of different dimensionality and sample sizes to compare the approach presented here with other algorithms. Real medical diagnoses and a biological classification of mushrooms are used to illustrate the performance of the algorithm. These results confirm the effectiveness of the proposed algorithm.Publication Metadata only Lexical acquisition and clustering of word senses to conceptual lexicon construction(2009-05) Pluempitiwiriyawej, C.; Cercone, N.; An, X.; Mahidol University. Faculty of Information and Communication TechnologyWe describe a mechanism and an algorithm to support construction of a large complex conceptual lexicon from an existing alphabetical lexicon. As part of this research, we define lexical models to present words and lexicons. Given the fact that an alphabetical lexicon contains lexical information about words which are organized by their spelling, constructing a conceptual lexicon requires an identification of lexical concepts and their relationships. Lexical acquisition and word-sense clustering are introduced to identify the lexical concepts and to discover the conceptual relationships. The result of this research is a set of candidate concepts which can be treated as initial concepts for the conceptual lexicon construction.Publication Metadata only Ontology-based multiperspective requirements traceability framework(2009) Assawamekin, N.; Sunetnanta, T.; Pluempitiwiriyawej, C.; ชาญยศ ปลื้มปิติวิริยะเวช; Mahidol University. Faculty of Information and Communication TechnologyLarge-scaled software development inevitably involves a group of stakeholders, each of whom may express their requirements differently in their own terminology and representation depending on their perspectives or perceptions of their shared problems. In view of that, the heterogeneity must be well handled and resolved in tracing and managing changes of such requirements. This paper presents our multiperspective requirements traceability (MUPRET) framework which deploys ontology as a knowledge management mechanism to intervene mutual "understanding" without restricting the freedom in expressing requirements differently. Ontology matching is applied as a reasoning mechanism in automatically generating traceability relationships. The relationships are identified by deriving semantic analogy of ontology concepts representing requirements elements. The precision and recall of traceability relationships generated by the framework are verified by comparing with a set of traceability relationships manually identified by users as a proof-of-concept of this framework.