2014 11th International Joint Conference on Computer Science and Software Engineering (JCSSE)最新文献

{"title":"Text corpus for natural language story-telling sentence generation: A design and evaluation","authors":"W. Limpanadusadee, P. Punyabukkana, A. Suchato, Onintra Poobrasert","doi":"10.1109/JCSSE.2014.6841846","DOIUrl":"https://doi.org/10.1109/JCSSE.2014.6841846","url":null,"abstract":"Automatic generation of narrative sentences from unordered word sets is desirable in Augmentative and Alternative Communication (AAC) systems for children with certain learning disabilities (LD). Regardless of the complexity of the Natural Language Processing deployed in sentence generation procedures, the qualities of language models always affect the generation results. This work compared sentence generation accuracies obtained from a multi-tier N-gram-based procedure trained on BEST2010, a large publicly available text corpus, and a smaller but more specifically designed corpus in the task of Thai simple sentence generation. The latter, a new corpus called TELL-S, was created based on an analysis of the contents belonging to textbooks used in grade 1 and grade 2 for Thai language subjects according to the compulsory curriculum for Thai schools. The original procedure was also modified to incorporate additional constraints based on a story-telling guideline developed for LD children. Evaluated upon test sets of 195 sentences, each of which was composed of 3-6 words with a specific Part-Of-Speech combination, TELL-S was shown to provide better generalization and yielded higher accuracies than BEST2010 in all cases with unbiased word sets. The sentence generation accuracies were 100% and 70% for 3-word and 4-word sentences, respectively. The average accuracy was at 58.8% when longer sentences were also included.","PeriodicalId":331610,"journal":{"name":"2014 11th International Joint Conference on Computer Science and Software Engineering (JCSSE)","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122537532","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}

{"title":"An improved adaptive discriminant analysis for single sample face recognition","authors":"Thitipan Wannakijmongkol, Ittiwat Khornrakhun, T. Chalidabhongse","doi":"10.1109/JCSSE.2014.6841833","DOIUrl":"https://doi.org/10.1109/JCSSE.2014.6841833","url":null,"abstract":"Face recognition is an automated process with the ability to identify individuals by their facial characteristics. Currently there is a problem in which the process requires several examples of the person of interest's face in order to produce accurate outcome, and the process is intolerant to the variation in facial expression and the condition of lighting of the face image needed to be identify. This inspired us to come up with an algorithm to increase accuracy of single sample facial recognition process. In the case where multiple samples are available, the best approach to identify a person by face recognition system is to use Fischer Linear Discriminant Analysis (FLDA) method which use multiple samples to calculate the within-class scatter matrix and could give output accurately. However with only one sample it means the sample does not have any variation, hence impossible to find the within-class scatter matrix. The Adaptive Discriminant Learning (ADL) [1] was proposed to solve the problem by deducing the within-class scatter matrix from auxiliary generic set which consist of multiple samples per person then use FLDA to recognize face image. In this paper, we improve the method by preprocessing the input image using a local illumination normalization to make the feature of the face became more obvious and suppress the effect of illumination variation and incorporating a part-based methodology to further increase the recognition rate. The system was tested with the FERET face database, and the recognition rate is improved from 77% to 93%.","PeriodicalId":331610,"journal":{"name":"2014 11th International Joint Conference on Computer Science and Software Engineering (JCSSE)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129149705","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}

{"title":"Parallel health tourism information extraction and ontology storage","authors":"Chidchanok Choksuchat, Suphaksa Ngamphak, Benjaporn Maneesaeng, Yuwathida Chiwpreechar, C. Chantrapornchai","doi":"10.1109/JCSSE.2014.6841873","DOIUrl":"https://doi.org/10.1109/JCSSE.2014.6841873","url":null,"abstract":"Health tourism is now popular and being promoted in Thailand since it is one of the growing industry. Health tourism information is scattered around in many places especially, in the websites. A health tourism service provider may be in many forms such as in the hotel, as a separated business, as a hospital etc. Each service provider may have its own website as well as the websites from common providers such as Tripadvisor, Agoda, or Atsiam, etc. Each website provides different information about just one service provider. In this work, we are interested in information gathering process for health tourism. We introduce the use of parallel Java platform to gather the tourism information particularly, using a Java concurrent program and merge the information using MapReduce. The information we gather are preprocessed and combined with the information manually collected. Google Refine is used to merge all the information into single health tourism ontology.","PeriodicalId":331610,"journal":{"name":"2014 11th International Joint Conference on Computer Science and Software Engineering (JCSSE)","volume":"121 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125435590","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}

{"title":"Prediction of tone naturalness perception using geometric model","authors":"K. Trakulsuk, A. Suchato, P. Punyabukkana, C. Wutiwiwatchai","doi":"10.1109/JCSSE.2014.6841845","DOIUrl":"https://doi.org/10.1109/JCSSE.2014.6841845","url":null,"abstract":"Naturalness is an important issue in the Text-To-Speech (TTS) system. To support arbitrarily defined pitch contours for any synthesized syllables, a TTS should be able to maintain the naturalness of the synthetic speech. This work proposed an automatic evaluation of pitch contours in order to determine the level of naturalness of synthesized syllables when perceived by human listeners. By analyzing results, tone perception experiments conducted on human listeners in this work, a syllable tone naturalness prediction model based on the midpoint and endpoint of the syllable's rhyme part was proposed. The model was then used for developing a tone naturalness prediction algorithm using geometric models of pitch contours. The evaluation of the tone naturalness prediction algorithm involved human listeners perceiving the naturalness of syllables with 45 pitch contour patterns, each of which with 2 repetitions. The proposed algorithm achieved approximately 80% consistency rate compared against human listeners' decisions on tone naturalness of the syllables.","PeriodicalId":331610,"journal":{"name":"2014 11th International Joint Conference on Computer Science and Software Engineering (JCSSE)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134213757","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}

{"title":"Analysis of factors which impact Facebook users' attitudes and behaviours using decision tree techniques","authors":"D. V. Hieu, N. Wisitpongphan, P. Meesad","doi":"10.1109/JCSSE.2014.6841880","DOIUrl":"https://doi.org/10.1109/JCSSE.2014.6841880","url":null,"abstract":"The number of Facebook users has increased dramatically over the past decades, as Facebook become the most common communication and sharing channel via the Internet. More people are now spending a lot of time using Facebook and gain high levels of satisfaction from using Facebook. Facebook users' attitudes and behaviours are becoming more influenced by highly technological devices which are not only integrated their personal lives but also areas of education and employment. This paper analyses factors including Internet and email use, owning a tablet, access to the Internet via mobile devices, gender, income, education, employment status. Moreover, an investigation of parental influence to Facebook users' attitudes and behaviours based on surveyed data from more than a thousand users using decision tree techniques.","PeriodicalId":331610,"journal":{"name":"2014 11th International Joint Conference on Computer Science and Software Engineering (JCSSE)","volume":"80 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117130587","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}

{"title":"Multi-camera based human localization for room utilization monitoring system","authors":"Issariyapol Siriwat, Thanaruk Theeramankong, Ithipan Methasate, H. Kunieda","doi":"10.1109/JCSSE.2014.6841840","DOIUrl":"https://doi.org/10.1109/JCSSE.2014.6841840","url":null,"abstract":"Room utilization monitoring system is an interesting application towards optimized facility usage. Due to its cost effectiveness and robustness, a camera-based system for room utilization is focused. This paper presents a multiple camera based process for room utilization monitoring system. The system is composed with three parts of processes that are single-camera processing, multi-camera processing and room event detection. In single camera processing, we have presented the object detection and tracking and transform the detected position on the image to the room map using Homography. The results of single camera processing are the location on the room map of detected object which are the input for multiple camera data fusion. We have purposed three different methods on multiple camera data fusion that are Uniform Bias Weighting, Best Camera Selection and Error Bias Weighting. And then we use the best detected object location on the map for room event detection. The experiment verified the performance of the different data fusion method on multiple cameras and showed that Error Bias weighting provide the best result.","PeriodicalId":331610,"journal":{"name":"2014 11th International Joint Conference on Computer Science and Software Engineering (JCSSE)","volume":"107 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124846757","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}

{"title":"An improvement of flat approach on hierarchical text classification using top-level pruning classifiers","authors":"Natchanon Phachongkitphiphat, P. Vateekul","doi":"10.1109/JCSSE.2014.6841847","DOIUrl":"https://doi.org/10.1109/JCSSE.2014.6841847","url":null,"abstract":"Hierarchical classification has been becoming a popular research topic nowadays, particularly on the web as text categorization. For a large web corpus, there can be a hierarchy with hundreds of thousands of topics, so it is common to handle this task using a flat classification approach, inducing a binary classifier only for the leaf-node classes. However, it always suffers from such low prediction accuracy due to an imbalanced issue in the training data. In this paper, we propose two novel strategies: (i) “Top-Level Pruning” to narrow down the candidate classes, and (ii) “Exclusive Top-Level Training Policy” to build more effective classifiers by utilizing the top-level data. The experiments on the Wikipedia dataset show that our system outperforms the traditional flat approach unanimously on all hierarchical classification metrics.","PeriodicalId":331610,"journal":{"name":"2014 11th International Joint Conference on Computer Science and Software Engineering (JCSSE)","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126307155","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}

{"title":"Modification of MELD score by including Serum Albumin to improve prediction of mortality outcome of cirrhotic patient based on Thai cirrhotic patients","authors":"Montri Duangkrut, Yaowadee Temtanapat, P. Komolmit","doi":"10.1109/JCSSE.2014.6841850","DOIUrl":"https://doi.org/10.1109/JCSSE.2014.6841850","url":null,"abstract":"Nowadays, the Model for End-stage Liver Disease (MELD) has become a popular model and replaced the Child-Pugh score for the assessment of the mortality opportunity of patients with cirrhosis in 3-month period. The model predicts the severity of the disease based on 3 biochemical parameters: serum creatinine, serum total bilirubin, and INR. However, in the past, the first model like Child-Pugh score signified the importance of Serum Albumin, a protein producing in a liver. It is, thus, expected that the Serum Albumin has an effect on patients' mortality prediction. In this research, our main focus is to refine and evaluate the effect of Serum Albumin to mortality of Thai cirrhotic patients if included into the MELD model. We use the data collection from 158 Thai cirrhotic patients with different degrees of severity. They were treated at the Liver Unit and Clinic, King Chulalongkorn Memorial Hospital, The Thai Red Cross Society. The collected data were divided into the periods of 3 months, 6 months, 1 year and 2 years respectively[1]. The Kaplan-Meier statistic was used to analyze the survival opportunity of each period. Also, the Cox-Regression was utilized to evaluate the relationship and the statistical significance of the substance in each period in order to find the connection between the Serum Albumin and mortality opportunities. Results of the study show that of all the data from 158 patients, with the Serum Albumin level between 1.0 and 3.5 g/dL, when tested by Pearson's Chi-squared[2], Log Rank Test and Wilcoxon rank-sum (Mann-Whitney)[3] has the statistical significance at the 1% level of confidence (p <; 0.001). Moreover, the correlation of the results using Cox Regression demonstrated also that Serum Albumin influenced the mortality opportunity at the hazard ratio of 5.14 (95%CI:2.971-8.920) with level of confidence p-value <; 0.0001. Thus, we believe that the Serum Albumin affected the mortality prediction model. We also propose two refined MELD models[4], ThaiMELD-Albumin and ThaiMELD-CTP[5]. For the efficiency assessment of the models, we compare our models to others using the ROC. We found that ThaiMELD-Albumin had 0.85 (95% CI: 0.68-1.00) and it is better than MELD, MELD-Albumin and 5vMELD, while ThaiMELD-CTP is just better than MELD. Consequently, ThaiMELD-Albumin is better for prediction of the mortality opportunity for Thai patients than the MELD, MELD-Albumin or 5vMELD. While ThaiMELD-CTP which just added a scale value to MELD could give a better assessment than MELD itself. Therefore, our model could benefit to Thai patients for the assessment of mortality opportunity as well as symptoms' severity. It could, perhaps, be further used for the consideration of liver transplantation in Thailand.","PeriodicalId":331610,"journal":{"name":"2014 11th International Joint Conference on Computer Science and Software Engineering (JCSSE)","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127749271","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}

{"title":"A novel segmentation method for isointense MRI brain tumor","authors":"Chaiyanan Sompong, S. Wongthanavasu","doi":"10.1109/JCSSE.2014.6841877","DOIUrl":"https://doi.org/10.1109/JCSSE.2014.6841877","url":null,"abstract":"This paper presents a novel segmentation method for isointense signal tumor appeared in T1-weighted or T2-weighted magnetic resonance (MR) images. The proposed method improves the well-known Grow-cut algorithm using the improved local transition rule. It applied the level set theory to extract tumor from the background by using strength probability surface map by threshold value. Heaviside step function are applied to assign the boundary among seed and background. For performance evaluation, tumor datasets on isointense signal with T1-weighted MRI acquired from Kitware/MIDAS repository are experimented throughout. The well-known grow-cut and tumorcut algorithms are compared using dice similarity coefficient (DSC). In this regard, the proposed method provides the better results by reporting DSC of 84.17 % higher than Grow-cut and Tumorcut with 80.81% and 80.14%, respectively.","PeriodicalId":331610,"journal":{"name":"2014 11th International Joint Conference on Computer Science and Software Engineering (JCSSE)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115360705","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}

{"title":"Vehicle logo detection using convolutional neural network and pyramid of histogram of oriented gradients","authors":"Wasin Thubsaeng, Aram Kawewong, K. Patanukhom","doi":"10.1109/JCSSE.2014.6841838","DOIUrl":"https://doi.org/10.1109/JCSSE.2014.6841838","url":null,"abstract":"This paper presents a new method for vehicle logo detection and recognition from images of front and back views of vehicle. The proposed method is a two-stage scheme which combines Convolutional Neural Network (CNN) and Pyramid of Histogram of Gradient (PHOG) features. CNN is applied as the first stage for candidate region detection and recognition of the vehicle logos. Then, PHOG with Support Vector Machine (SVM) classifier is employed in the second stage to verify the results from the first stage. Experiments are performed with dataset of vehicle images collected from internet. The results show that the proposed method can accurately locate and recognize the vehicle logos with higher robustness in comparison with the other conventional schemes. The proposed methods can provide up to 100% in recall, 96.96% in precision and 99.99% in recognition rate in dataset of 20 classes of the vehicle logo.","PeriodicalId":331610,"journal":{"name":"2014 11th International Joint Conference on Computer Science and Software Engineering (JCSSE)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124934645","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}