Pub Date : 2015-08-01DOI: 10.1109/ICAICTA.2015.7335375
D. Lestari, Angela Irfani
Performance of Indonesian Automatic Speech Recognition is decreased significantly when recognizing spontaneous speech. Spontaneous speech has particular characteristics differ from read speech both in acoustic and language rule. In spontaneous speech, the pronunciation and expression of the speech varies depending on the speaker fluency and the topic. Disfluencies in speech disrupt a fluent sentence and more often violates the rule of the formal language. To improve Indonesian automatic speech recognizer to recognize spontaneous speech, several model enhancement methods was conducted by adding spontaneous data and retrain both acoustic model and language model using those data, by adapting the acoustic model based on the maximum likelihood linear regression and maximum a posteriori approach, and by adapting the language model employing the language model linear interpolation. Experimental results show all methods are effective in increasing the capability of the Indonesian automatic speech recognizer to recognize spontaneous data. However, all methods decreased the accuracy of read speech recognition. On average, retraining both acoustic and language models using combination of read and spontaneous data is more effective than conducting model adaptation. The absolute improvement of 28.34% accuracy is achieved after retraining both language model and acoustic model using combination of read data and spontaneous data.
{"title":"Acoustic and language models adaptation for Indonesian spontaneous speech recognition","authors":"D. Lestari, Angela Irfani","doi":"10.1109/ICAICTA.2015.7335375","DOIUrl":"https://doi.org/10.1109/ICAICTA.2015.7335375","url":null,"abstract":"Performance of Indonesian Automatic Speech Recognition is decreased significantly when recognizing spontaneous speech. Spontaneous speech has particular characteristics differ from read speech both in acoustic and language rule. In spontaneous speech, the pronunciation and expression of the speech varies depending on the speaker fluency and the topic. Disfluencies in speech disrupt a fluent sentence and more often violates the rule of the formal language. To improve Indonesian automatic speech recognizer to recognize spontaneous speech, several model enhancement methods was conducted by adding spontaneous data and retrain both acoustic model and language model using those data, by adapting the acoustic model based on the maximum likelihood linear regression and maximum a posteriori approach, and by adapting the language model employing the language model linear interpolation. Experimental results show all methods are effective in increasing the capability of the Indonesian automatic speech recognizer to recognize spontaneous data. However, all methods decreased the accuracy of read speech recognition. On average, retraining both acoustic and language models using combination of read and spontaneous data is more effective than conducting model adaptation. The absolute improvement of 28.34% accuracy is achieved after retraining both language model and acoustic model using combination of read data and spontaneous data.","PeriodicalId":319020,"journal":{"name":"2015 2nd International Conference on Advanced Informatics: Concepts, Theory and Applications (ICAICTA)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130993441","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 : 2015-08-01DOI: 10.1109/ICAICTA.2015.7335379
Dody Dharma, Afwarman Manaf
In this paper, we explain about the implementation of interactive fluid simulation based on Navier-Stokes equations for mobile devices. We use Material Point Method (Euler-Lagrange) approach for fluid discretization. This implementation is optimized to reach 60 cycles/second of simulation. The dynamic computing optimization is done by using several techniques, such as: grid size reduction, quadratic B-spline interpolation, and localized active node computation. The gravitational direction is determined by data stream from accelerometer sensor. The external forces of fluid are generated from touch screen sensor and its coordinate. We also propose a strategy for fluid visualization in Graphic Processing Unit (GPU) in order to reach rendering performance at 60 frame-per-second (fps) by using motion blur technique. From this research, we obtain 2000 particles that can be simulated interactively at 60 fps.
{"title":"Interactive fluid simulation based on material point method for mobile devices","authors":"Dody Dharma, Afwarman Manaf","doi":"10.1109/ICAICTA.2015.7335379","DOIUrl":"https://doi.org/10.1109/ICAICTA.2015.7335379","url":null,"abstract":"In this paper, we explain about the implementation of interactive fluid simulation based on Navier-Stokes equations for mobile devices. We use Material Point Method (Euler-Lagrange) approach for fluid discretization. This implementation is optimized to reach 60 cycles/second of simulation. The dynamic computing optimization is done by using several techniques, such as: grid size reduction, quadratic B-spline interpolation, and localized active node computation. The gravitational direction is determined by data stream from accelerometer sensor. The external forces of fluid are generated from touch screen sensor and its coordinate. We also propose a strategy for fluid visualization in Graphic Processing Unit (GPU) in order to reach rendering performance at 60 frame-per-second (fps) by using motion blur technique. From this research, we obtain 2000 particles that can be simulated interactively at 60 fps.","PeriodicalId":319020,"journal":{"name":"2015 2nd International Conference on Advanced Informatics: Concepts, Theory and Applications (ICAICTA)","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131717626","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 : 2015-08-01DOI: 10.1109/ICAICTA.2015.7335365
M. L. Khodra
Event extraction identifies who did what, when, where, why, and how, which is known as 5W1H. We aim to investigate event extraction on Indonesian news articles as multiclass-categorization problem, and apply statistical learning-based approach that treats event extraction as a sequence labeling problem under BIO (Begin Inside Outside) labeling scheme. Each token of input text will be classified into one of 13 predefined classes. Our contributions are providing 5W1H corpus, and the best technique to build model of event extraction. Our experiments show that C4.5 is better than AdaboostM1 although Adaboost can identify minority labels better than C4.5. In addition, C4.5 with all features gave the best Fmeasure of 0.666.
{"title":"Event extraction on Indonesian news article using multiclass categorization","authors":"M. L. Khodra","doi":"10.1109/ICAICTA.2015.7335365","DOIUrl":"https://doi.org/10.1109/ICAICTA.2015.7335365","url":null,"abstract":"Event extraction identifies who did what, when, where, why, and how, which is known as 5W1H. We aim to investigate event extraction on Indonesian news articles as multiclass-categorization problem, and apply statistical learning-based approach that treats event extraction as a sequence labeling problem under BIO (Begin Inside Outside) labeling scheme. Each token of input text will be classified into one of 13 predefined classes. Our contributions are providing 5W1H corpus, and the best technique to build model of event extraction. Our experiments show that C4.5 is better than AdaboostM1 although Adaboost can identify minority labels better than C4.5. In addition, C4.5 with all features gave the best Fmeasure of 0.666.","PeriodicalId":319020,"journal":{"name":"2015 2nd International Conference on Advanced Informatics: Concepts, Theory and Applications (ICAICTA)","volume":"103 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131488163","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 : 2015-08-01DOI: 10.1109/ICAICTA.2015.7335374
Sounak Paul, Tapan Dey
Wireless Sensor Network (WSN) is severely energy constrained. Selecting an energy efficient routing is very important for such network. In this paper, we present an approach for routing in wireless sensor network to minimize the energy consumption to increase active lifetime of the network. Our approach takes the minimum energy route for transmitting data from a node to a fixed base station through a cluster head. We calculate the distance between neighboring clusters using Hausdorff distance. We evaluate energy consumed to transmit a `k' bit message to all the neighbors from source cluster. Comparing the entire route we select the minimum energy route to transmit the data to the base station. Simulation result shows that our scheme achieves significant improvement in terms of energy efficiency when compared with some of the established routing protocols in the same category.
{"title":"Energy efficient routing in cluster based wireless sensor networks","authors":"Sounak Paul, Tapan Dey","doi":"10.1109/ICAICTA.2015.7335374","DOIUrl":"https://doi.org/10.1109/ICAICTA.2015.7335374","url":null,"abstract":"Wireless Sensor Network (WSN) is severely energy constrained. Selecting an energy efficient routing is very important for such network. In this paper, we present an approach for routing in wireless sensor network to minimize the energy consumption to increase active lifetime of the network. Our approach takes the minimum energy route for transmitting data from a node to a fixed base station through a cluster head. We calculate the distance between neighboring clusters using Hausdorff distance. We evaluate energy consumed to transmit a `k' bit message to all the neighbors from source cluster. Comparing the entire route we select the minimum energy route to transmit the data to the base station. Simulation result shows that our scheme achieves significant improvement in terms of energy efficiency when compared with some of the established routing protocols in the same category.","PeriodicalId":319020,"journal":{"name":"2015 2nd International Conference on Advanced Informatics: Concepts, Theory and Applications (ICAICTA)","volume":"50 1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121209865","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 : 2015-08-01DOI: 10.1109/ICAICTA.2015.7335353
Abu Nowshed Chy, Md Zia Ullah, Masaki Aono
Microblog, especially Twitter, have become an integral part of our daily life for searching latest news and events information. Due to short length characteristics of tweets, only content-relevance based search result cannot satisfy user's information need. Recent research shows that considering temporal aspects in this regard improve the retrieval performance significantly. In this paper, we propose a method to re-rank the search result based on temporal features, account related features, and Twitter specific features along with textual features of tweets. We also applied a two stage query expansion technique to improve the relevancy of tweets. After automatic feature selection by using LASSO and elastic-net regularization; we applied random forest as a feature ranking method to estimate the importance of selected feature. Then, with that importance score, a weighted ranking model combines the features value to estimate the relevance score. We conducted our experiments based on the TREC Microblog 2011 and 2012 queries over the TREC Tweets2011 collection. Experimental result demonstrates the effectiveness of our method over the baseline in terms of precision@30 (P@30), mean average precision (MAP), and reciprocal-precision (R-Prec) metrics.
{"title":"Combining temporal and content aware features for microblog retrieval","authors":"Abu Nowshed Chy, Md Zia Ullah, Masaki Aono","doi":"10.1109/ICAICTA.2015.7335353","DOIUrl":"https://doi.org/10.1109/ICAICTA.2015.7335353","url":null,"abstract":"Microblog, especially Twitter, have become an integral part of our daily life for searching latest news and events information. Due to short length characteristics of tweets, only content-relevance based search result cannot satisfy user's information need. Recent research shows that considering temporal aspects in this regard improve the retrieval performance significantly. In this paper, we propose a method to re-rank the search result based on temporal features, account related features, and Twitter specific features along with textual features of tweets. We also applied a two stage query expansion technique to improve the relevancy of tweets. After automatic feature selection by using LASSO and elastic-net regularization; we applied random forest as a feature ranking method to estimate the importance of selected feature. Then, with that importance score, a weighted ranking model combines the features value to estimate the relevance score. We conducted our experiments based on the TREC Microblog 2011 and 2012 queries over the TREC Tweets2011 collection. Experimental result demonstrates the effectiveness of our method over the baseline in terms of precision@30 (P@30), mean average precision (MAP), and reciprocal-precision (R-Prec) metrics.","PeriodicalId":319020,"journal":{"name":"2015 2nd International Conference on Advanced Informatics: Concepts, Theory and Applications (ICAICTA)","volume":"470 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116413266","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 : 2015-08-01DOI: 10.1109/ICAICTA.2015.7335350
V. Patanavijit
In the research operation of Digital Signal Processing (DSP) and Digital Image Processing (DIP), one of the most essential obstacles is the image denoise algorithm by the reason of a very large demand of high quality noise-free images therefore there are many image denoise algorithms have been invented in the time of two decades. Bilateral filter is one of the most impressive and feasible algorithms, which is usually applied for denoise propose, but the performance of the Bilateral filter is substantially bank on three parameters: spatial variance, radiometric variance and window size. Consequently, this paper investigates the performance influence impact of spatial variance, radiometric variance, window size for the Bilateral Filter in the denoise propose. In the denoise experiment, Bilateral filter (BF) is applied on three noisy standard images under five Gaussian noise power levels and the best results in the PSNR prospective point of view from deniose algorithm is picked. Moreover, an optimal value of three parameters: spatial variance, radiometric variance, window size, which make the performance of Bilateral filter the highest PSNR, are extensively investigated for each types of tested images and each noise powers.
{"title":"The bilateral denoising performance influence of window, spatial and radiometric variance","authors":"V. Patanavijit","doi":"10.1109/ICAICTA.2015.7335350","DOIUrl":"https://doi.org/10.1109/ICAICTA.2015.7335350","url":null,"abstract":"In the research operation of Digital Signal Processing (DSP) and Digital Image Processing (DIP), one of the most essential obstacles is the image denoise algorithm by the reason of a very large demand of high quality noise-free images therefore there are many image denoise algorithms have been invented in the time of two decades. Bilateral filter is one of the most impressive and feasible algorithms, which is usually applied for denoise propose, but the performance of the Bilateral filter is substantially bank on three parameters: spatial variance, radiometric variance and window size. Consequently, this paper investigates the performance influence impact of spatial variance, radiometric variance, window size for the Bilateral Filter in the denoise propose. In the denoise experiment, Bilateral filter (BF) is applied on three noisy standard images under five Gaussian noise power levels and the best results in the PSNR prospective point of view from deniose algorithm is picked. Moreover, an optimal value of three parameters: spatial variance, radiometric variance, window size, which make the performance of Bilateral filter the highest PSNR, are extensively investigated for each types of tested images and each noise powers.","PeriodicalId":319020,"journal":{"name":"2015 2nd International Conference on Advanced Informatics: Concepts, Theory and Applications (ICAICTA)","volume":"139 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115974366","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 : 2015-08-01DOI: 10.1109/ICAICTA.2015.7335363
N. Maulidevi, M. Al-Hasan, M. L. Khodra
Diesel Engine is still required in Indonesia, since diesel engine is the most appropriate engine to produce electricity in isolated area. Indonesia comprises of a large number of isolated area, and at the moment it is not the priority of the government to install cable connecting those isolated areas. The diesel engines can be used in the next 15-20 years, if the maintenance is properly conducted. The maintenance requires the online monitoring and appropriate recommendation based on the result of monitoring process. This paper discusses the integration of online monitoring system, which is discussed in the previous paper, and the proposed recommendation system. The online monitoring system is implemented as web based system, the recommendation is utilizing CLIPS as the back-end processes, and the integration is implemented in Java as web based application.
{"title":"Integration of Diesel Engine monitoring system and recommendation system","authors":"N. Maulidevi, M. Al-Hasan, M. L. Khodra","doi":"10.1109/ICAICTA.2015.7335363","DOIUrl":"https://doi.org/10.1109/ICAICTA.2015.7335363","url":null,"abstract":"Diesel Engine is still required in Indonesia, since diesel engine is the most appropriate engine to produce electricity in isolated area. Indonesia comprises of a large number of isolated area, and at the moment it is not the priority of the government to install cable connecting those isolated areas. The diesel engines can be used in the next 15-20 years, if the maintenance is properly conducted. The maintenance requires the online monitoring and appropriate recommendation based on the result of monitoring process. This paper discusses the integration of online monitoring system, which is discussed in the previous paper, and the proposed recommendation system. The online monitoring system is implemented as web based system, the recommendation is utilizing CLIPS as the back-end processes, and the integration is implemented in Java as web based application.","PeriodicalId":319020,"journal":{"name":"2015 2nd International Conference on Advanced Informatics: Concepts, Theory and Applications (ICAICTA)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132729343","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 : 2015-08-01DOI: 10.1109/ICAICTA.2015.7335362
Jittavat Suksumlarn, W. Suwannik, M. Maleewong
This paper presents the Lattice Boltzmann method (LBM) for two dimensional shallow water equations to simulate dam break problem. We implemented sequential code and compared it with the exact solution. Then, we implemented three parallel programs with the Compute Unified Device Architecture (CUDA). The first parallel version is a straightforward implementation. The second parallel version reduces the calculations of distribution function in the LBM. The third parallel program has fixed the branch divergence problem with branch distribution. Speed up for the third parallel program is increased approximately 2.1x from the first parallel program, and it is increased approximately 22.7x from the sequential program version.
{"title":"Lattice Boltzmann method for two-dimensional shallow water equations with CUDA","authors":"Jittavat Suksumlarn, W. Suwannik, M. Maleewong","doi":"10.1109/ICAICTA.2015.7335362","DOIUrl":"https://doi.org/10.1109/ICAICTA.2015.7335362","url":null,"abstract":"This paper presents the Lattice Boltzmann method (LBM) for two dimensional shallow water equations to simulate dam break problem. We implemented sequential code and compared it with the exact solution. Then, we implemented three parallel programs with the Compute Unified Device Architecture (CUDA). The first parallel version is a straightforward implementation. The second parallel version reduces the calculations of distribution function in the LBM. The third parallel program has fixed the branch divergence problem with branch distribution. Speed up for the third parallel program is increased approximately 2.1x from the first parallel program, and it is increased approximately 22.7x from the sequential program version.","PeriodicalId":319020,"journal":{"name":"2015 2nd International Conference on Advanced Informatics: Concepts, Theory and Applications (ICAICTA)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130499192","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 : 2015-08-01DOI: 10.1109/ICAICTA.2015.7335361
Phat Jotikabukkana, Virach Sornlertlamvanich, Okumura Manabu, C. Haruechaiyasak
Social media text can illustrate significant information of our real social situation. It can show the direction of real-time social movement. However, it has its own characteristics such as using short text and informal language, many unstructured information and argot. This kind of text is hard to classify and difficult to analyze to extract the useful information. In this paper, we propose an effective technique to classify the social media text by utilizing the initial keywords from well-formed sources of data, such as online news. Term frequency-inverse document frequency weighting technique (TF-IDF) and Word Article Matrix (WAM) are used as main methods in this research. We use the extracted keywords from the well-formed source as a main factor to do experiment on Twitter messages. We found a set of the social media keywords can represent the essence of social events and can be used to classify the text effectively.
{"title":"Effectiveness of social media text classification by utilizing the online news category","authors":"Phat Jotikabukkana, Virach Sornlertlamvanich, Okumura Manabu, C. Haruechaiyasak","doi":"10.1109/ICAICTA.2015.7335361","DOIUrl":"https://doi.org/10.1109/ICAICTA.2015.7335361","url":null,"abstract":"Social media text can illustrate significant information of our real social situation. It can show the direction of real-time social movement. However, it has its own characteristics such as using short text and informal language, many unstructured information and argot. This kind of text is hard to classify and difficult to analyze to extract the useful information. In this paper, we propose an effective technique to classify the social media text by utilizing the initial keywords from well-formed sources of data, such as online news. Term frequency-inverse document frequency weighting technique (TF-IDF) and Word Article Matrix (WAM) are used as main methods in this research. We use the extracted keywords from the well-formed source as a main factor to do experiment on Twitter messages. We found a set of the social media keywords can represent the essence of social events and can be used to classify the text effectively.","PeriodicalId":319020,"journal":{"name":"2015 2nd International Conference on Advanced Informatics: Concepts, Theory and Applications (ICAICTA)","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116880094","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 : 2015-08-01DOI: 10.1109/ICAICTA.2015.7335371
W. Maharani, D. H. Widyantoro, M. L. Khodra
Aspect extraction is an important task in sentiment analysis to identify aspects in customer review products. Most existing works defines the pattern set manually or using heuristic approach. In this paper, we propose SAE, a Syntactical-based Aspect Extraction using decision tree and rule learning to generate the pattern set based on sequence labelling. We provide a comprehensive analysis of aspect extraction using pattern-based method and typed-dependency. The patterns will be used to identify and extract aspect term candidates in customer product review. First, we generate pattern set that identify aspect term candidates using decision tree and rule learning such as ID3, J48, Random Tree, Part and Prism, based on sequence labelling. The set of pattern is employed to produced aspect term candidates. We use a list of positive and negative opinion lexicon as aspect term candidates filtering. Finally, we combine the pattern-based method with typed dependency to remove irrelevant aspect term. The results showed that the combination of pattern-based and typed dependency can increase the performance. However, since our work is based on syntactic-based approach, it can be used to other domains, that is expected to include an unlimited domain datasets.
{"title":"SAE: Syntactic-based aspect and opinion extraction from product reviews","authors":"W. Maharani, D. H. Widyantoro, M. L. Khodra","doi":"10.1109/ICAICTA.2015.7335371","DOIUrl":"https://doi.org/10.1109/ICAICTA.2015.7335371","url":null,"abstract":"Aspect extraction is an important task in sentiment analysis to identify aspects in customer review products. Most existing works defines the pattern set manually or using heuristic approach. In this paper, we propose SAE, a Syntactical-based Aspect Extraction using decision tree and rule learning to generate the pattern set based on sequence labelling. We provide a comprehensive analysis of aspect extraction using pattern-based method and typed-dependency. The patterns will be used to identify and extract aspect term candidates in customer product review. First, we generate pattern set that identify aspect term candidates using decision tree and rule learning such as ID3, J48, Random Tree, Part and Prism, based on sequence labelling. The set of pattern is employed to produced aspect term candidates. We use a list of positive and negative opinion lexicon as aspect term candidates filtering. Finally, we combine the pattern-based method with typed dependency to remove irrelevant aspect term. The results showed that the combination of pattern-based and typed dependency can increase the performance. However, since our work is based on syntactic-based approach, it can be used to other domains, that is expected to include an unlimited domain datasets.","PeriodicalId":319020,"journal":{"name":"2015 2nd International Conference on Advanced Informatics: Concepts, Theory and Applications (ICAICTA)","volume":"104 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132404496","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：