Pub Date : 2017-07-12DOI: 10.1109/JCSSE.2017.8025903
Boonyarit Deewattananon, Usa Sammapun
As more and more Thais own mobile devices, mobile applications are high in demand. Before installing mobile applications, many users read reviews written by other users to determine whether or not the application is worth using. In addition, mobile application developers also rely on user reviews to get insight information on which aspects of the mobile application users like or do not like and why. They can use the information to market the beloved aspects of their software product and improve on the problematic ones. However, when there are many reviews, it is difficult to comprehend information in the user reviews. Several researches in recent years aim to extract opinions and sentiments from various texts or documents such as Twitter, webboards, and software product reviews. Most of these researches are for English documents. For Thai language, researches usually focus on other contexts such as hotel reviews or general opinions on Twitter. In this paper, we present an approach to analyze user reviews written in Thai based on techniques in natural language processing, topic modeling, and sentiment analysis. The approach aims to help Thai users and developers discover dynamically, instead of pre-determined, various aspects and associated sentiments from a vast amount of user reviews. The result of the approach is a list of aspects with associated opinions and sentiments to help users assess mobile applications and provide summarized user feedbacks for developers.
{"title":"Analyzing user reviews in Thai language toward aspects in mobile applications","authors":"Boonyarit Deewattananon, Usa Sammapun","doi":"10.1109/JCSSE.2017.8025903","DOIUrl":"https://doi.org/10.1109/JCSSE.2017.8025903","url":null,"abstract":"As more and more Thais own mobile devices, mobile applications are high in demand. Before installing mobile applications, many users read reviews written by other users to determine whether or not the application is worth using. In addition, mobile application developers also rely on user reviews to get insight information on which aspects of the mobile application users like or do not like and why. They can use the information to market the beloved aspects of their software product and improve on the problematic ones. However, when there are many reviews, it is difficult to comprehend information in the user reviews. Several researches in recent years aim to extract opinions and sentiments from various texts or documents such as Twitter, webboards, and software product reviews. Most of these researches are for English documents. For Thai language, researches usually focus on other contexts such as hotel reviews or general opinions on Twitter. In this paper, we present an approach to analyze user reviews written in Thai based on techniques in natural language processing, topic modeling, and sentiment analysis. The approach aims to help Thai users and developers discover dynamically, instead of pre-determined, various aspects and associated sentiments from a vast amount of user reviews. The result of the approach is a list of aspects with associated opinions and sentiments to help users assess mobile applications and provide summarized user feedbacks for developers.","PeriodicalId":6460,"journal":{"name":"2017 14th International Joint Conference on Computer Science and Software Engineering (JCSSE)","volume":"101 1","pages":"1-6"},"PeriodicalIF":0.0,"publicationDate":"2017-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77424054","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2017-07-12DOI: 10.1109/JCSSE.2017.8025911
Ma. Shiela C. Sapul, T. Aung, Rachsuda Jiamthapthaksin
There is no previous research that compares the results of k-means, CLOPE clustering and Latent Dirichlet Allocation (LDA) topic modeling algorithms for detecting trending topics on tweets. Since not all tweets contain hashtags, we considered three training data feature sets: hashtags, keywords and keywords + hashtags in this study. Our proposed methodology proved that CLOPE can also be used in a non-transactional database like Twitter data set to answer the trending topic discovery and could provide more topic patterns than k-means and LDA. Using additional feature sets has improved the results of k-means and LDA, thus, keywords + hashtags can identify more meaningful topics.
{"title":"Trending topic discovery of Twitter Tweets using clustering and topic modeling algorithms","authors":"Ma. Shiela C. Sapul, T. Aung, Rachsuda Jiamthapthaksin","doi":"10.1109/JCSSE.2017.8025911","DOIUrl":"https://doi.org/10.1109/JCSSE.2017.8025911","url":null,"abstract":"There is no previous research that compares the results of k-means, CLOPE clustering and Latent Dirichlet Allocation (LDA) topic modeling algorithms for detecting trending topics on tweets. Since not all tweets contain hashtags, we considered three training data feature sets: hashtags, keywords and keywords + hashtags in this study. Our proposed methodology proved that CLOPE can also be used in a non-transactional database like Twitter data set to answer the trending topic discovery and could provide more topic patterns than k-means and LDA. Using additional feature sets has improved the results of k-means and LDA, thus, keywords + hashtags can identify more meaningful topics.","PeriodicalId":6460,"journal":{"name":"2017 14th International Joint Conference on Computer Science and Software Engineering (JCSSE)","volume":"9 1","pages":"1-6"},"PeriodicalIF":0.0,"publicationDate":"2017-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83430847","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2017-07-12DOI: 10.1109/JCSSE.2017.8025898
Narongsak Sukma, R. Chokngamwong
Caller ID has been used to tell the recipient who is calling before answering the call. In fact, nowadays using just Caller ID is not enough to proof the real caller since there are several ways to manipulate the caller identity. There are number of solutions to proof the caller e.g. using Time base, SMS base, or hardware. Even using DSA and CA, it can lead to data leak or inconsistent verification processing. The One-time password practices can mitigate the risk of Man-in-the-middle attacks because SSL has vulnerability assessment that can lead to MITM or man in the middle attack. The attacker can intercept SSL verification process between Server and client for sniffing then spoofing. It would be better if we can find a solution that does not rely on CA, Third party and/or external hardware. In this paper, we propose the solution with self-controlled security and one-time key issue to avoid data leak. The one-time key issuance is a good solution for verification and detecting caller ID Spoofing attacker through this methodology since it does not rely on third-party CA and store certification anywhere. This solution provides the best of key management as the one-time secret key is used. Results from our test lab show effectively verification rates and good performance where resource and power consumption are not impacted.
{"title":"One time key Issuing for Verification and Detecting Caller ID Spoofing Attacks","authors":"Narongsak Sukma, R. Chokngamwong","doi":"10.1109/JCSSE.2017.8025898","DOIUrl":"https://doi.org/10.1109/JCSSE.2017.8025898","url":null,"abstract":"Caller ID has been used to tell the recipient who is calling before answering the call. In fact, nowadays using just Caller ID is not enough to proof the real caller since there are several ways to manipulate the caller identity. There are number of solutions to proof the caller e.g. using Time base, SMS base, or hardware. Even using DSA and CA, it can lead to data leak or inconsistent verification processing. The One-time password practices can mitigate the risk of Man-in-the-middle attacks because SSL has vulnerability assessment that can lead to MITM or man in the middle attack. The attacker can intercept SSL verification process between Server and client for sniffing then spoofing. It would be better if we can find a solution that does not rely on CA, Third party and/or external hardware. In this paper, we propose the solution with self-controlled security and one-time key issue to avoid data leak. The one-time key issuance is a good solution for verification and detecting caller ID Spoofing attacker through this methodology since it does not rely on third-party CA and store certification anywhere. This solution provides the best of key management as the one-time secret key is used. Results from our test lab show effectively verification rates and good performance where resource and power consumption are not impacted.","PeriodicalId":6460,"journal":{"name":"2017 14th International Joint Conference on Computer Science and Software Engineering (JCSSE)","volume":"77 1","pages":"1-4"},"PeriodicalIF":0.0,"publicationDate":"2017-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81127723","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2017-07-12DOI: 10.1109/JCSSE.2017.8025922
Dilok Pumsuwan, S. Rimcharoen, Nutthanon Leelathakul
Crossover methods are important keys to the success of genetic algorithms. However, traditional crossover methods fail to solve a trap problem, which is a difficult benchmark problem designed to deceive genetic algorithms to favor all-zero bits, while the actual solution is all-one bits. The Bayesian optimization algorithm (BOA) is the most famous algorithm that can solve the trap problem; however, it incurs a large computational cost. This paper, therefore, proposes a novel crossover technique, called a front-rear crossover (FRC), to enhance the simple genetic algorithm. We test the proposed technique with various benchmark problems and compare the results with four other crossover algorithms, including single point crossover (SPC), two point crossover (TPC), uniform crossover (UC) and ring crossover (RC). The FRC outperforms the four techniques in all test problems. It can also solve the trap problem by requiring the 40 times lesser number of fitness evaluations than BOA's.
{"title":"Front-rear crossover: A new crossover technique for solving a trap problem","authors":"Dilok Pumsuwan, S. Rimcharoen, Nutthanon Leelathakul","doi":"10.1109/JCSSE.2017.8025922","DOIUrl":"https://doi.org/10.1109/JCSSE.2017.8025922","url":null,"abstract":"Crossover methods are important keys to the success of genetic algorithms. However, traditional crossover methods fail to solve a trap problem, which is a difficult benchmark problem designed to deceive genetic algorithms to favor all-zero bits, while the actual solution is all-one bits. The Bayesian optimization algorithm (BOA) is the most famous algorithm that can solve the trap problem; however, it incurs a large computational cost. This paper, therefore, proposes a novel crossover technique, called a front-rear crossover (FRC), to enhance the simple genetic algorithm. We test the proposed technique with various benchmark problems and compare the results with four other crossover algorithms, including single point crossover (SPC), two point crossover (TPC), uniform crossover (UC) and ring crossover (RC). The FRC outperforms the four techniques in all test problems. It can also solve the trap problem by requiring the 40 times lesser number of fitness evaluations than BOA's.","PeriodicalId":6460,"journal":{"name":"2017 14th International Joint Conference on Computer Science and Software Engineering (JCSSE)","volume":"5 1","pages":"1-6"},"PeriodicalIF":0.0,"publicationDate":"2017-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79452045","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2017-07-01DOI: 10.1109/JCSSE.2017.8025900
Korn Poonsirivong, C. Jittawiriyanukoon
Anomaly detection (outlier) using simulation helps us analyze the anomaly instances from big data source. As the hasty explosion of today's data stream, outlier detection technique will be an analytical tool to be employed for evaluating massive unstructured datasets. In order to speed-up the processing time to handle enormous datasets, this research will conduct experiments of advanced distant-based outlier detection algorithms to investigate the most effective algorithms using MOA. The algorithms used in this study are Continuous Outlie Detection (COD), Micro-Cluster based COD or MCOD, and STream OutlierR Miner (STORM). The results demonstrate MCOD algorithm can outperform other two algorithms in terms of processing time and accurate anomalies.
{"title":"A rapid anomaly detection technique for big data curation","authors":"Korn Poonsirivong, C. Jittawiriyanukoon","doi":"10.1109/JCSSE.2017.8025900","DOIUrl":"https://doi.org/10.1109/JCSSE.2017.8025900","url":null,"abstract":"Anomaly detection (outlier) using simulation helps us analyze the anomaly instances from big data source. As the hasty explosion of today's data stream, outlier detection technique will be an analytical tool to be employed for evaluating massive unstructured datasets. In order to speed-up the processing time to handle enormous datasets, this research will conduct experiments of advanced distant-based outlier detection algorithms to investigate the most effective algorithms using MOA. The algorithms used in this study are Continuous Outlie Detection (COD), Micro-Cluster based COD or MCOD, and STream OutlierR Miner (STORM). The results demonstrate MCOD algorithm can outperform other two algorithms in terms of processing time and accurate anomalies.","PeriodicalId":6460,"journal":{"name":"2017 14th International Joint Conference on Computer Science and Software Engineering (JCSSE)","volume":"47 1","pages":"1-6"},"PeriodicalIF":0.0,"publicationDate":"2017-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76550378","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Nowadays, recorded videos from surveillance cameras are important evidence for legal investigation in the field of forensic science. Videos may be modified to deviate contents by a person involves in a crime. In this paper, a video editing detection based on Scalable Color Descriptor (SCD) and Color Layout Descriptor (CLD) is proposed. The detection method is composed of two components: (1) generating video identifier and signature and (2) video verification. The experimental results show that applying SCD and CLD to design the detection method outperforms the other descriptors in terms of false acceptance rate and false rejection rate. It is concluded that our method accurately classifies whether or not an incoming video is forged.
{"title":"Robust video editing detection using Scalable Color and Color Layout Descriptors","authors":"Peerapon Chantharainthron, Sasipa Panthuwadeethorn, Suphakant Phimoltares","doi":"10.1109/JCSSE.2017.8025923","DOIUrl":"https://doi.org/10.1109/JCSSE.2017.8025923","url":null,"abstract":"Nowadays, recorded videos from surveillance cameras are important evidence for legal investigation in the field of forensic science. Videos may be modified to deviate contents by a person involves in a crime. In this paper, a video editing detection based on Scalable Color Descriptor (SCD) and Color Layout Descriptor (CLD) is proposed. The detection method is composed of two components: (1) generating video identifier and signature and (2) video verification. The experimental results show that applying SCD and CLD to design the detection method outperforms the other descriptors in terms of false acceptance rate and false rejection rate. It is concluded that our method accurately classifies whether or not an incoming video is forged.","PeriodicalId":6460,"journal":{"name":"2017 14th International Joint Conference on Computer Science and Software Engineering (JCSSE)","volume":"32 1","pages":"1-6"},"PeriodicalIF":0.0,"publicationDate":"2017-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80067315","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2017-07-01DOI: 10.1109/JCSSE.2017.8025936
Matee Charoendee, A. Suchato, P. Punyabukkana
Speech emotion recognition is a challenging problem, with identifying efficient features being of particular concern. This paper has two components. First, it presents an empirical study that evaluated four feature reduction methods, chi-square, gain ratio, RELIEF-F, and kernel principal component analysis (KPCA), on utterance level using a support vector machine (SVM) as a classifier. KPCA had the highest F-score when its F-score was compared with the average F-score of the other methods. Using KPCA is more effective than classifying without using feature reduction methods up to 5.73%. The paper also presents an application of statistical functions to raw features from the segment level to derive global features. The features were then reduced using KPCA and classified with SVM. Subsequently, we conducted a majority vote to determine the emotion for the entire utterance. The results demonstrate that this approach outperformed the baseline approaches, which used features from the utterance level, the utterance level with KPCA, the segment level, the segment level with KPCA, and the segment level with the application of statistical functions without KPCA. This yielded a higher F-score at 13.16%, 7.03%, 5.13%, 4.92% and 11.04%, respectively.
{"title":"Speech emotion recognition using derived features from speech segment and kernel principal component analysis","authors":"Matee Charoendee, A. Suchato, P. Punyabukkana","doi":"10.1109/JCSSE.2017.8025936","DOIUrl":"https://doi.org/10.1109/JCSSE.2017.8025936","url":null,"abstract":"Speech emotion recognition is a challenging problem, with identifying efficient features being of particular concern. This paper has two components. First, it presents an empirical study that evaluated four feature reduction methods, chi-square, gain ratio, RELIEF-F, and kernel principal component analysis (KPCA), on utterance level using a support vector machine (SVM) as a classifier. KPCA had the highest F-score when its F-score was compared with the average F-score of the other methods. Using KPCA is more effective than classifying without using feature reduction methods up to 5.73%. The paper also presents an application of statistical functions to raw features from the segment level to derive global features. The features were then reduced using KPCA and classified with SVM. Subsequently, we conducted a majority vote to determine the emotion for the entire utterance. The results demonstrate that this approach outperformed the baseline approaches, which used features from the utterance level, the utterance level with KPCA, the segment level, the segment level with KPCA, and the segment level with the application of statistical functions without KPCA. This yielded a higher F-score at 13.16%, 7.03%, 5.13%, 4.92% and 11.04%, respectively.","PeriodicalId":6460,"journal":{"name":"2017 14th International Joint Conference on Computer Science and Software Engineering (JCSSE)","volume":"8 1","pages":"1-6"},"PeriodicalIF":0.0,"publicationDate":"2017-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"78032322","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2017-07-01DOI: 10.1109/JCSSE.2017.8025935
Arkane Khaminkure, Paramate Horkaew, J. Panyavaraporn
Brain atlas has become a primary means of computer aided neurological diagnosis. It relies on registering intra/inter-subject brain scans on a common frame of reference, on which statistical variability model is built. This diffeomorphic map of anatomically plausible correspondence could in turn be used for monitoring and identifying progress and manifestation of the disease, respectively. It is accepted that dense image registration is very accurate but computationally expensive. This paper thus presents a feature based image registration by using orientation invariant Gabor responses of texture. The reported results herein demonstrate that it is both anatomically accurate and robust.
{"title":"Building a brain atlas based on gabor texture features","authors":"Arkane Khaminkure, Paramate Horkaew, J. Panyavaraporn","doi":"10.1109/JCSSE.2017.8025935","DOIUrl":"https://doi.org/10.1109/JCSSE.2017.8025935","url":null,"abstract":"Brain atlas has become a primary means of computer aided neurological diagnosis. It relies on registering intra/inter-subject brain scans on a common frame of reference, on which statistical variability model is built. This diffeomorphic map of anatomically plausible correspondence could in turn be used for monitoring and identifying progress and manifestation of the disease, respectively. It is accepted that dense image registration is very accurate but computationally expensive. This paper thus presents a feature based image registration by using orientation invariant Gabor responses of texture. The reported results herein demonstrate that it is both anatomically accurate and robust.","PeriodicalId":6460,"journal":{"name":"2017 14th International Joint Conference on Computer Science and Software Engineering (JCSSE)","volume":"46 1","pages":"1-5"},"PeriodicalIF":0.0,"publicationDate":"2017-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73179708","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2017-07-01DOI: 10.1109/JCSSE.2017.8025926
Naladtaporn Aottiwerch, Urachart KoKaew
Microbial science is inevitably involved in human daily life because these creatures live around us. They are both useful and punishable. If the people have knowledge about microorganisms, it can control the microbes that cause the wicked things, and it can also be useful to use microorganisms effectively. Currently, the study still use the microscope in the education. There are limitations when exploring various cycles, it will not be able to explore as the time required. As the cycles of nature, it must have time as a variable. Therefore, the microscope helps to see the structure clearer only, but it cannot see all the natural cycles covered by a single microscope. Therefore, this research uses the Augmented Reality technology to be developed as instructional media (A case study of Phylum Basidiomycota, the fungi, which is considered to be the most complex internal structure). This helps to visualize the three-dimensional structure and cycles of nature from electronic devices. But the problem of instructional media created with Augmented Reality technology is that it is inaccessible and difficult to use. The researcher has developed together with online instructional media for easy access, and designed instructional media with the theory of Shneiderman's 8 Golden Rules for ease of use, and measured the performance with real applications by the third-year students of the Faculty of Science, Khon kaen University. It was found that Augmented Reality instructional media in online system was accessible and easy to use. It is also very satisfying for users, with an average score of 4.5 out of 1–5, this means that the criteria is very good.
微生物科学不可避免地与人类的日常生活有关，因为这些生物生活在我们周围。它们既有用又要受到惩罚。如果人们对微生物有所了解，就可以控制导致邪恶事物的微生物，也可以有效地利用微生物。目前，在教育研究中仍使用显微镜。在探索各种循环时存在局限性，它将无法按照所需的时间进行探索。正如自然界的周期一样，它必须有时间作为变量。因此，显微镜只能帮助更清楚地看到结构，而不能看到单个显微镜覆盖的所有自然循环。因此，本研究采用增强现实技术作为教学媒介(以担子菌门为例，担子菌门被认为是内部结构最复杂的真菌)。这有助于从电子设备中可视化三维结构和自然界的循环。但利用增强现实技术创建的教学媒体存在着难以接近和难以使用的问题。研究者与在线教学媒体共同开发了易于访问的教学媒体，并以Shneiderman’s 8 Golden Rules理论设计了易于使用的教学媒体，并通过实际应用对孔原大学理学院三年级学生的表现进行了测量。发现在线系统中的增强现实教学媒体具有可访问性和易用性。对于用户来说，它也非常满意，在1-5分中平均得分为4.5分，这意味着这个标准非常好。
{"title":"Design computer-assisted learning in an online Augmented Reality environment based on Shneiderman's eight Golden Rules","authors":"Naladtaporn Aottiwerch, Urachart KoKaew","doi":"10.1109/JCSSE.2017.8025926","DOIUrl":"https://doi.org/10.1109/JCSSE.2017.8025926","url":null,"abstract":"Microbial science is inevitably involved in human daily life because these creatures live around us. They are both useful and punishable. If the people have knowledge about microorganisms, it can control the microbes that cause the wicked things, and it can also be useful to use microorganisms effectively. Currently, the study still use the microscope in the education. There are limitations when exploring various cycles, it will not be able to explore as the time required. As the cycles of nature, it must have time as a variable. Therefore, the microscope helps to see the structure clearer only, but it cannot see all the natural cycles covered by a single microscope. Therefore, this research uses the Augmented Reality technology to be developed as instructional media (A case study of Phylum Basidiomycota, the fungi, which is considered to be the most complex internal structure). This helps to visualize the three-dimensional structure and cycles of nature from electronic devices. But the problem of instructional media created with Augmented Reality technology is that it is inaccessible and difficult to use. The researcher has developed together with online instructional media for easy access, and designed instructional media with the theory of Shneiderman's 8 Golden Rules for ease of use, and measured the performance with real applications by the third-year students of the Faculty of Science, Khon kaen University. It was found that Augmented Reality instructional media in online system was accessible and easy to use. It is also very satisfying for users, with an average score of 4.5 out of 1–5, this means that the criteria is very good.","PeriodicalId":6460,"journal":{"name":"2017 14th International Joint Conference on Computer Science and Software Engineering (JCSSE)","volume":"45 1","pages":"1-5"},"PeriodicalIF":0.0,"publicationDate":"2017-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82625625","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2017-07-01DOI: 10.1109/JCSSE.2017.8025934
M. Phyu, Kiyota Hashimoto
Word segmentation is one of the most fundamental processes for most natural language processing tasks. In particular, languages with no word boundary in writing such as Chinese, Japanese, Korean, Thai, and Burmese need it. However, the Burmese language still waits for a technique with good performance. In this paper, we propose a new technique for Burmese word segmentation employing the idea of Character Clustering for Conditional Random Fields. Character clusters are groups of some inseparable characters due to language characteristics. We proposed a set of 29 types of Burmese Character Clusters (BCCs) as rules, and Conditional Random Fields is applied as a sequential labelling machine learning method. We compared our proposed method with CRF without BCC and Syllable-based CRFs. The result shows that our proposed method achieved the highest performance.
分词是大多数自然语言处理任务中最基本的过程之一。特别是，汉语、日语、韩语、泰语、缅甸语等没有文字边界的语言需要它。然而，缅甸语仍在等待一种表现良好的技术。本文提出了一种基于条件随机场的字符聚类思想的缅甸语分词新技术。字簇是由于语言的特点而形成的一组不可分割的字。我们提出了一组29种缅甸语字符簇(bcc)作为规则，并将条件随机场(Conditional Random Fields)作为顺序标记机器学习方法。我们将该方法与不含BCC的CRF和基于音节的CRF进行了比较。结果表明，本文提出的方法达到了最高的性能。
{"title":"Burmese word segmentation with Character Clustering and CRFs","authors":"M. Phyu, Kiyota Hashimoto","doi":"10.1109/JCSSE.2017.8025934","DOIUrl":"https://doi.org/10.1109/JCSSE.2017.8025934","url":null,"abstract":"Word segmentation is one of the most fundamental processes for most natural language processing tasks. In particular, languages with no word boundary in writing such as Chinese, Japanese, Korean, Thai, and Burmese need it. However, the Burmese language still waits for a technique with good performance. In this paper, we propose a new technique for Burmese word segmentation employing the idea of Character Clustering for Conditional Random Fields. Character clusters are groups of some inseparable characters due to language characteristics. We proposed a set of 29 types of Burmese Character Clusters (BCCs) as rules, and Conditional Random Fields is applied as a sequential labelling machine learning method. We compared our proposed method with CRF without BCC and Syllable-based CRFs. The result shows that our proposed method achieved the highest performance.","PeriodicalId":6460,"journal":{"name":"2017 14th International Joint Conference on Computer Science and Software Engineering (JCSSE)","volume":"23 1","pages":"1-6"},"PeriodicalIF":0.0,"publicationDate":"2017-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87374280","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：