Ekkarat Boonchieng

Training set size reduction in large dataset problems

Classifiers have known to be used in various fields of applications. However, the main problem usually found recently is about applying a classifier to large datasets. Thus, the process of reducing size of the training set becomes necessary especially to accelerate the processing time of the classifier. Concerning the problem, this paper proposes a new method which can reduce size of the training set in a large dataset. Our proposed method is improved from a famous graph-based algorithm named Optimum-Path Forest (OPF). Our principal concept of reducing the training sets size is to utilize the Segmented Least Square Algorithm (SLSA) in estimating the trees shape. From the experimental results, our proposed method could reduce size of the training set by about 7 to 21 percent comparing with the original OPF algorithm while the classifications accuracy decreased insignificantly by only about 0.2 to 0.5 percent. In addition, for some datasets, our method provided even as same degree of accuracy as of the original OPF algorithm.

IOT for smart farm: A case study of the Lingzhi mushroom farm at Maejo University

This research aims to prototype a smart Lingzhi mushroom farm. This research applied the use of IOT with a sensor to measure and monitors the humidity in the Lingzhi mushroom farm. The humidity data processed through NETPIE was developed and provided by NECTEC as a free service for IOT. Humidity data was stored into a NET FEED (a sub service from NETPIE) and displayed on mobile devices and computers through NET FREEBOARD (another sub service of NETPIE). This research also controlled sprinkler and fog pumps automatically and the functional status (switching on and off for periods of time) pushes notifications through LINE API on the LINE Application. The equipment and tools used in this research were NodeMCU, humidity sensor, RTC (real time clock), relay module, sprinkler and fog pumps. C++ and Node.JS were used as programming. The services and protocol used were NETPIE (Network Platform for internet of everything) with subservices such as NETPIE FEED, NETPIE FREEBOARD, and NETPIE REST API. The results of the research showed that using IOT with the sensor enhanced the prototype of smart farming.

An integrated system of applying the use of Internet of Things, RFID and cloud computing: A case study of logistic management of Electricity Generation Authority of Thailand (EGAT) Mae Mao Lignite Coal Mining, Lampang, Thailand

This research aims to find the best practice of logistic management for Electricity Generation Authority of Thailand (EGAT) Mae Mao Mining, Lampang. This research applied the use of RFID for lignite coal trucks and data from RFID proceed through a server and was stored into a private cloud computing. The equipment and tools used in research was a RFID reader, UHF passive RFID tags, Arduino Mega 2560 + Ethernet Shield, PHP, Jason, Node.JS and Maria DB as a database system. The protocol used was MQTT. 27 trucks had UHF passive tags installed, 1 crusher (location number 3) had an RFID reader installed. This system has been operating 24 hours a day 7 days a week, from the beginning of 2015 to mid 2016. The results of the research show that officers who worked for related systems were satisfied. The system enhanced the best practice of lignite coal mining logistic in terms of information checking.

Efficiency comparisons between k-centers and k-means algorithms

This paper proposes an under-sampling method with an algorithm which guarantees the sampling quality called k-centers algorithm. Then, the efficiency of the sampling using under-sampling method with k-means algorithm is compared with the proposed method. For the comparison purpose, four datasets obtained from UCI database were selected and the RIPPER classifier was used. From the experimental results, our under-sampling method with k-centers algorithm provided the Accuracy, Recall, and F-measure values higher than that obtained from the under-sampling with k-means algorithm in every dataset we used. The Precision value from our k-centers algorithm might be lower in some datasets, however, its average value computed out of all datasets is still higher than using the under-sampling method with k-means algorithm. Moreover, the experimental results showed that our under-sampling method with k-centers algorithm also decreases the Accuracy value obtained from the original data less than that using the under-sampling with k-means algorithm.

Graphical representation of the whole sequentially MRI images in a single view image sequences of human's whole head

Typically, a sequence of the Magnetic Resonance Imaging (MRI) images is composed of a certain number of images projecting some internal organs of a human, such as brain and eyeballs of the humans head, which is the case chosen for our demonstration. Each of MRI images in the sequence presents only a thin layer of the whole head. The image processing techniques proposed in this paper aims to allow all such sequential images to be visible through a single view. In other words, the whole head of a human can be visible in just one image and thus looks as a three-dimensional view of the head. Unfortunately, there may be some deviation in positions even between contiguous images in the sequence. Centroid of the humans head appeared in each image should be measured. To ensure a centroids position is estimated well enough so that centroids of all sequential images are not so much deviated from each other, searching for the centroid of a humans whole head is done based on an approximate convex shape rather than a circular shape as usual. From our experimental results, there is no significant deviation of centroids between contiguous frames as expected.

Brain tumor's approximate correspondence and area with interior holes filled

Measuring area of tumor in humans brain from only single image may provide incorrect information for further diagnosis. Generally, a doctor or an expert must examine a brain tumor from several sequential MRI images to conclude its size or the severity level of patients illness. To imitate the way a doctor diagnosing such case in a real situation, some digital image processing techniques are proposed and applied in order to provide support for a tentative or an initial analysis to the doctor. Thus, correspondence of appearances of a tumor presented in all MRI images should be linked and considered. In image processing, a closed area can be seen as an object and based on the similarity of its interior shadings, the objects centroid can be estimated. Unfortunately, although an objects centroid may be calculated even there exists slightly different shadings which are still considered as having similarity inside the closed shape of the object, only a small hole can cause deviation of computed centroid from its expected position. Since the typical thresholding techniques still leave a hole whose area has a certain amount of different shading from the major shading of the objects area. Thus, we proposed a number of image processing techniques for the purpose of tumor area approximation. Moreover, the proposed methods include a correspondence technique would also support multiple-object detection and linking centroids of the same object, which is a brain tumor in this case, presented in a pair of contiguous images.

Digital disease detection: Application of machine learning in community health informatics

Health informatics is a new research area which is interdisciplinary amongst information science, computer science and healthcare. The concept of health informatics is to develop a new way to manipulate healthcare data from various resources and devices by optimizing the method of data acquisition, data storage, data processing, and data visualization. Community health informatics can be described as the systematic application of information and computer science to obtain valuable data for solving health problems and providing it to health policy makers. The challenge of community health informatics is to maximize the efficiency and efficacy of big data analysis. This discussion paper aims to present the various applications of machine learning and software engineering approaches that implemented in digital disease detection.

Development of thai word segmentation technique for solving problems with unknown words

This research has an objective to develop an efficient technique for Thai word segmentation, especially those nonexistent in dictionaries. The researchers developed a model for Thai word segmentation by relying on grammar and rules to solve the problems with words not found in dictionaries. The model was intended to be used as the best approach of word segmentation, which applied the segmentation technique developed by the researchers called PTTSF (Parsing Thai Text with Syntax and Feature of Word). The system of this technique operates by starting from finding the boundary of each word in Thai sentences. If the system finds a word that does not exist in the dictionary or a meaningless word, it would not be able to solve the problem with the method of longest-matching algorithm. Therefore, rules need to be specified to solve such problems. In this study, 28 rules were created and Digraph method was used to find a pattern of word segmentation with the highest probability based on the grammatical principle. After the procedure of finding boundary of the word, the result from correct word segmentation can be used for further processes. In analyzing efficiency of the system, its accuracy in word segmentation was the main point of concern. The results revealed that the derived mapping technique could solve the problem concerned with segmentation words that do not exist in the dictionary with an average accuracy over 90% of the whole document. However, the researchers encountered with ambiguous words problem. Although this problem rarely occurs, it could affect accuracy of word segmentation.