Pub Date : 2020-06-01DOI: 10.1109/ECTI-CON49241.2020.9158060
Phattaleeya Mabpa, Piyasawat Navaratana Na Ayudhya, J. Kunthong
Clogged sewer pipelines are one of the main problems that cause Sanitary Sewer Overflow ( SSO) which leads to serious environmental issues and property damage. This work presented clogged pipe detection and monitoring methods based on acoustic analysis to identify pipe clogged occurrence and degree of blockage that can be mitigating the risks from SSO’s problems. The technique is to attach the vibration speaker on the pipe as an acoustic source. The clogged by blockage will be detected by reading the change in the pipe resonance frequency via the microphone installed on the other side of the pipeline. The resonance frequency of the measured signals was characterized by Fast Fourier Transforms (FFT). Compensation based line was used to normalize the frequency responses for easier acoustic analysis. The experiments have indicated the resonance frequency shifting down and Sound Pressure Level (SPL) decreasing when pipe clogged. Moreover, this paper provided the technique that offers small and noncomplex data which any classification method can identify pipe clogged easier in future work.
{"title":"Clogged Pipe Detection and Monitoring by Using Acoustic Analysis Methodology","authors":"Phattaleeya Mabpa, Piyasawat Navaratana Na Ayudhya, J. Kunthong","doi":"10.1109/ECTI-CON49241.2020.9158060","DOIUrl":"https://doi.org/10.1109/ECTI-CON49241.2020.9158060","url":null,"abstract":"Clogged sewer pipelines are one of the main problems that cause Sanitary Sewer Overflow ( SSO) which leads to serious environmental issues and property damage. This work presented clogged pipe detection and monitoring methods based on acoustic analysis to identify pipe clogged occurrence and degree of blockage that can be mitigating the risks from SSO’s problems. The technique is to attach the vibration speaker on the pipe as an acoustic source. The clogged by blockage will be detected by reading the change in the pipe resonance frequency via the microphone installed on the other side of the pipeline. The resonance frequency of the measured signals was characterized by Fast Fourier Transforms (FFT). Compensation based line was used to normalize the frequency responses for easier acoustic analysis. The experiments have indicated the resonance frequency shifting down and Sound Pressure Level (SPL) decreasing when pipe clogged. Moreover, this paper provided the technique that offers small and noncomplex data which any classification method can identify pipe clogged easier in future work.","PeriodicalId":371552,"journal":{"name":"2020 17th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology (ECTI-CON)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131566080","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}
Myocardial Ischemia is the main cause of mortality in patients with Coronary Artery Disease (CAD). One of the methods used in screening patients with this disease is the diag-nosis of radionuclide myocardial perfusion imaging (rMPI). In this paper, we conducted a comparative study by experimenting on several machine learning models, such as Logistic Regression, Random Forest, XGBoost, etc., to classify the stenosis of coronary artery. High-level features from rMPI computed by 4D-MSPECT polar map were used to train/test the models. rMPI features of the risk group of CAD patients were obtained from a public hospital. With the hypothesis that patient characteristics (e.g., Diabetes Mellitus, Hypertension, Dyslipidemia) could improve the prediction performance of the models, this study also included patient characteristics in our experimentation as important parts of feature selection. All other processes (i.e., data cleaning, feature selection, feature engineering and feature transformation) in machine learning pipeline were also deliberately experimented in this study. For model selection, two-level validation regarding generalization and hyperparameter tuning were also performed.
{"title":"Comparing Classifiers for the Prediction of the Stenosis of Coronary Artery","authors":"Hataichanok Aakkara, Atumporn Aaisueb, Aeerapong Aeelanupab","doi":"10.1109/ecti-con49241.2020.9158312","DOIUrl":"https://doi.org/10.1109/ecti-con49241.2020.9158312","url":null,"abstract":"Myocardial Ischemia is the main cause of mortality in patients with Coronary Artery Disease (CAD). One of the methods used in screening patients with this disease is the diag-nosis of radionuclide myocardial perfusion imaging (rMPI). In this paper, we conducted a comparative study by experimenting on several machine learning models, such as Logistic Regression, Random Forest, XGBoost, etc., to classify the stenosis of coronary artery. High-level features from rMPI computed by 4D-MSPECT polar map were used to train/test the models. rMPI features of the risk group of CAD patients were obtained from a public hospital. With the hypothesis that patient characteristics (e.g., Diabetes Mellitus, Hypertension, Dyslipidemia) could improve the prediction performance of the models, this study also included patient characteristics in our experimentation as important parts of feature selection. All other processes (i.e., data cleaning, feature selection, feature engineering and feature transformation) in machine learning pipeline were also deliberately experimented in this study. For model selection, two-level validation regarding generalization and hyperparameter tuning were also performed.","PeriodicalId":371552,"journal":{"name":"2020 17th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology (ECTI-CON)","volume":"52 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133405056","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 : 2020-06-01DOI: 10.1109/ECTI-CON49241.2020.9157906
Sathaporn Konjunthes, W. Thaiwirot, P. Akkaraekthalin
This paper presents a wideband circularly polarized stacked patch antenna with truncated corners ground plane for universal UHF RFID reader. The configuration of the antenna composes of a slanted elliptical patch, corner truncated patch, a horizontally meandered strip (HMS), and corner truncated ground plane. One end of the HMS is connected to an SMA connector, while the other end is connected to the radiating patch via probe. In order to improve the bandwidth, the parasitic patch is used by placing above the radiating patch. From the simulated results, the proposed antenna has an impedance bandwidth of about 18.6% (800 – 965 MHz), a 3-dB axial ratio (AR) bandwidth of around 18.3% (816 – 981 MHz), and maximum gain of 8.4 dBi. The proposed antenna can operate in the frequency range of 840 – 960 MHz. Therefore, it is suitable for applying in universal UHF RFID reader.
{"title":"A Wideband Circularly Polarized Stacked Patch Antenna With Truncated Corners Ground Plane for Universal UHF RFID Reader","authors":"Sathaporn Konjunthes, W. Thaiwirot, P. Akkaraekthalin","doi":"10.1109/ECTI-CON49241.2020.9157906","DOIUrl":"https://doi.org/10.1109/ECTI-CON49241.2020.9157906","url":null,"abstract":"This paper presents a wideband circularly polarized stacked patch antenna with truncated corners ground plane for universal UHF RFID reader. The configuration of the antenna composes of a slanted elliptical patch, corner truncated patch, a horizontally meandered strip (HMS), and corner truncated ground plane. One end of the HMS is connected to an SMA connector, while the other end is connected to the radiating patch via probe. In order to improve the bandwidth, the parasitic patch is used by placing above the radiating patch. From the simulated results, the proposed antenna has an impedance bandwidth of about 18.6% (800 – 965 MHz), a 3-dB axial ratio (AR) bandwidth of around 18.3% (816 – 981 MHz), and maximum gain of 8.4 dBi. The proposed antenna can operate in the frequency range of 840 – 960 MHz. Therefore, it is suitable for applying in universal UHF RFID reader.","PeriodicalId":371552,"journal":{"name":"2020 17th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology (ECTI-CON)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117173939","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 : 2020-06-01DOI: 10.1109/ecti-con49241.2020.9158252
K. Phaebua, A. Boonpoonga, S. Burintramart
This paper proposes a study of a time-domain ground penetrating radar (GPR) signal by using the time- domain uniform geometrical theory of diffraction (UTD) method. The monostatic GPR system is employed in this study. Rays tracing method is employed to find the reflected ray and diffracted ray paths between the transmitting antenna (Tx) and the receiving antenna (Rx). Moreover, the conventional UTD method is adapted to calculate the frequency-domain transfer function of the electric field between Tx antenna and the Rx antenna. The medium including air, ground and underground objects are taken into account. Finally, the received time-domain GPR signal is obtained by using the inverse fast Fourier transform (iFFT) method. The calculated results found that the time-domain GPR signal can be calculated by using the time-domain UTD method. The ground reflection, object reflection and object diffraction time-domain signal can be separately calculated. The reflected signal from the ground and an underground curved object can be illustrated.
{"title":"Time-Domain GPR Signal Prediction by Using Time-Domain UTD Method","authors":"K. Phaebua, A. Boonpoonga, S. Burintramart","doi":"10.1109/ecti-con49241.2020.9158252","DOIUrl":"https://doi.org/10.1109/ecti-con49241.2020.9158252","url":null,"abstract":"This paper proposes a study of a time-domain ground penetrating radar (GPR) signal by using the time- domain uniform geometrical theory of diffraction (UTD) method. The monostatic GPR system is employed in this study. Rays tracing method is employed to find the reflected ray and diffracted ray paths between the transmitting antenna (Tx) and the receiving antenna (Rx). Moreover, the conventional UTD method is adapted to calculate the frequency-domain transfer function of the electric field between Tx antenna and the Rx antenna. The medium including air, ground and underground objects are taken into account. Finally, the received time-domain GPR signal is obtained by using the inverse fast Fourier transform (iFFT) method. The calculated results found that the time-domain GPR signal can be calculated by using the time-domain UTD method. The ground reflection, object reflection and object diffraction time-domain signal can be separately calculated. The reflected signal from the ground and an underground curved object can be illustrated.","PeriodicalId":371552,"journal":{"name":"2020 17th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology (ECTI-CON)","volume":"112 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117287515","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 : 2020-06-01DOI: 10.1109/ecti-con49241.2020.9158238
Nutcha Likhitkitwoerakul, Natchanai Roongmuanpha, W. Tangsrirat
This paper presents the floating impedance function simulator using flipped voltage follower-based differential voltage to current converter (DVTC) along with three passive elements. With the proper selection of the three passive components, the floating inductor, capacitor and frequency- dependent negative resistor (FDNR) can be realized. The realized equivalent impedance value of the proposed simulator can be controlled conveniently by changing the passive element values. A second-order bandstop filter and third-order lowpass filter are designed using the proposed simulators as application examples. To verify the workability of the proposed circuit, simulation results are confirmed through PSPICE program using 0.25-μm CMOS process parameters from TSMC (Taiwan Semiconductor Manufacturing Company).
{"title":"Floating Impedance Simulator Realization","authors":"Nutcha Likhitkitwoerakul, Natchanai Roongmuanpha, W. Tangsrirat","doi":"10.1109/ecti-con49241.2020.9158238","DOIUrl":"https://doi.org/10.1109/ecti-con49241.2020.9158238","url":null,"abstract":"This paper presents the floating impedance function simulator using flipped voltage follower-based differential voltage to current converter (DVTC) along with three passive elements. With the proper selection of the three passive components, the floating inductor, capacitor and frequency- dependent negative resistor (FDNR) can be realized. The realized equivalent impedance value of the proposed simulator can be controlled conveniently by changing the passive element values. A second-order bandstop filter and third-order lowpass filter are designed using the proposed simulators as application examples. To verify the workability of the proposed circuit, simulation results are confirmed through PSPICE program using 0.25-μm CMOS process parameters from TSMC (Taiwan Semiconductor Manufacturing Company).","PeriodicalId":371552,"journal":{"name":"2020 17th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology (ECTI-CON)","volume":"95 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115002693","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 : 2020-06-01DOI: 10.1109/ECTI-CON49241.2020.9158094
Yuzana Hlaing, Nyein Aye Maung Maung
Time-based wireless indoor localization recently stands as the state-of-the-art situation for the up-to-date real time application areas. Synchronizing time among the wireless nodes may still be the major challenging problem for time- based indoor localization on wireless networks. The main reason of this problem is the difficulty and complexity of using standard time synchronization protocols such as Network Time Protocol (NTP), and Coordinated Universal Time (UTC) on wireless networks. Besides, the limitations of the wireless networks restrict the consideration of which time synchronization protocols should be applied on which wireless communication technologies such as WiFi, LiFi, Bluetooth, UWB, Ultrasonic, ZigBee, etc, for wireless time synchronization. Generally, time synchronization schemes support better accuracy if and only if precise random delay model can be implemented for the estimation of random delay caused by the real environmental impacts. In this paper, a ZigBee-based hybrid wireless time synchronization approach, which is required as the first and vital step for time-based wireless indoor localization, is proposed. In order to provide precise synchronization accuracy of the proposed hybrid approach to be more precise, appropriate random delay model is implemented by applying the empirical approach on ZigBee- based test-bed. By using the empirical data obtained from anchor-to-anchor packet communications, random delay is estimated using Gaussian random delay model and then least square regression model is applied to improve synchronization accuracy. In accordance with the experimental evaluation results, it highlights that synchronization performance of the proposed approach is significantly improved.
{"title":"Hybrid Time Synchronization for ZigBee Networks: An Empirical Approach","authors":"Yuzana Hlaing, Nyein Aye Maung Maung","doi":"10.1109/ECTI-CON49241.2020.9158094","DOIUrl":"https://doi.org/10.1109/ECTI-CON49241.2020.9158094","url":null,"abstract":"Time-based wireless indoor localization recently stands as the state-of-the-art situation for the up-to-date real time application areas. Synchronizing time among the wireless nodes may still be the major challenging problem for time- based indoor localization on wireless networks. The main reason of this problem is the difficulty and complexity of using standard time synchronization protocols such as Network Time Protocol (NTP), and Coordinated Universal Time (UTC) on wireless networks. Besides, the limitations of the wireless networks restrict the consideration of which time synchronization protocols should be applied on which wireless communication technologies such as WiFi, LiFi, Bluetooth, UWB, Ultrasonic, ZigBee, etc, for wireless time synchronization. Generally, time synchronization schemes support better accuracy if and only if precise random delay model can be implemented for the estimation of random delay caused by the real environmental impacts. In this paper, a ZigBee-based hybrid wireless time synchronization approach, which is required as the first and vital step for time-based wireless indoor localization, is proposed. In order to provide precise synchronization accuracy of the proposed hybrid approach to be more precise, appropriate random delay model is implemented by applying the empirical approach on ZigBee- based test-bed. By using the empirical data obtained from anchor-to-anchor packet communications, random delay is estimated using Gaussian random delay model and then least square regression model is applied to improve synchronization accuracy. In accordance with the experimental evaluation results, it highlights that synchronization performance of the proposed approach is significantly improved.","PeriodicalId":371552,"journal":{"name":"2020 17th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology (ECTI-CON)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115092587","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 : 2020-06-01DOI: 10.1109/ecti-con49241.2020.9158081
Sansit Ngamrassameewong, Vichaya Manatchinapisit, Y. Wongsawat
Human expression are exhibited throughout by language, which is generally used for communications. Making use of this fact, combining the language expression silently with the imagination of movement may lead to the improvement on developing the brain-computer interface (BCI) system. This paper focuses on the silent speech related to upper limbs movement. In the experiment, five subjects performed silent speech, i.e. at the same time, imagined upper limbs movement follow by two words, which are "RIGHT" or "LEFT". Furthermore, each of the participants performed another experiment doing only motor imagery (MI). In both experiments, there are five trials for each movement side. 19-channel EEG was recorded. The EEG signals were simply analyzed by using the absolute power. The result shows the clearer activation of event-related desynchronization (ERD) of the alpha band (8-12 Hz) on the motor area (C3, C4) during the combination of silent speech and MI.
{"title":"Improvement of Motor Imagery BCI using Silent Speech","authors":"Sansit Ngamrassameewong, Vichaya Manatchinapisit, Y. Wongsawat","doi":"10.1109/ecti-con49241.2020.9158081","DOIUrl":"https://doi.org/10.1109/ecti-con49241.2020.9158081","url":null,"abstract":"Human expression are exhibited throughout by language, which is generally used for communications. Making use of this fact, combining the language expression silently with the imagination of movement may lead to the improvement on developing the brain-computer interface (BCI) system. This paper focuses on the silent speech related to upper limbs movement. In the experiment, five subjects performed silent speech, i.e. at the same time, imagined upper limbs movement follow by two words, which are \"RIGHT\" or \"LEFT\". Furthermore, each of the participants performed another experiment doing only motor imagery (MI). In both experiments, there are five trials for each movement side. 19-channel EEG was recorded. The EEG signals were simply analyzed by using the absolute power. The result shows the clearer activation of event-related desynchronization (ERD) of the alpha band (8-12 Hz) on the motor area (C3, C4) during the combination of silent speech and MI.","PeriodicalId":371552,"journal":{"name":"2020 17th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology (ECTI-CON)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115457117","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}
This paper presents the study of lead acid battery for charging process using Hydrogen (H) and Oxygen (O2) gas release condition. The battery of 12V/100 Ah is selected to study from the purposed. The battery charging methodology is combined from the basically charging with constant current, constant voltage and gas analysis method for finding fully charging condition. The measurement data from the charging process show gas leases with relate the H and O2 gas varies from the battery voltage. The O2 and H gas release can be defined into 2 sections. The second section the O2 gas release has high changed in more than H gas release that point the H gas will be linearized trend. Theses point of the second section can define the fully charging process of the LA battery with 15.5 V of battery voltage. Therefore, this method can be defined as maximum voltage for charging the LA battery and adapted in the future.
{"title":"Novel Battery Charging Method using Hydrogen and Oxygen Gas Release Condition for Lead Acid Battery","authors":"Nirutti Nilkeaw, Chairat Sornchai, Boonyang Plungklang","doi":"10.1109/ecti-con49241.2020.9158269","DOIUrl":"https://doi.org/10.1109/ecti-con49241.2020.9158269","url":null,"abstract":"This paper presents the study of lead acid battery for charging process using Hydrogen (H) and Oxygen (O2) gas release condition. The battery of 12V/100 Ah is selected to study from the purposed. The battery charging methodology is combined from the basically charging with constant current, constant voltage and gas analysis method for finding fully charging condition. The measurement data from the charging process show gas leases with relate the H and O2 gas varies from the battery voltage. The O2 and H gas release can be defined into 2 sections. The second section the O2 gas release has high changed in more than H gas release that point the H gas will be linearized trend. Theses point of the second section can define the fully charging process of the LA battery with 15.5 V of battery voltage. Therefore, this method can be defined as maximum voltage for charging the LA battery and adapted in the future.","PeriodicalId":371552,"journal":{"name":"2020 17th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology (ECTI-CON)","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116651925","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}
Topic modeling is an unsupervised learning approach, which can automatically discover the hidden thematic structure in text documents. For text mining, topic modeling is a language-independent technique that disregards grammar and word order. Apart from semantic and structural issues, Thai language is typically considered more complex than others. Due to the lack of word delimiter and a surfeit of composite words. Errors from word tokenization can create significant problems for any post processes of text, such as document retrieval, sentiment analysis, machine translation, etc., adversely decreasing the performance of text applications. Despite a strong correlation between word ordering and semantic meaning, topic modeling has been widely reported that it can extract latent information, aka. latent topic or latent semantic, encoded in documents. Although there were few previous research works on studying topic modeling in Thai language, they mostly focused on upstream processes of Natural Language Processing (NLP) in, for example, applying a refined stop-word list to, or adding N-gram on a single specific topic modeling method. To our knowledge, this paper is the first to explore different topic modeling approaches, i.e., Latent Dirichlet Allocation (LDA) and Nonnegative Metrix Factorization (NMF), in Thai Language to compare their coherence. We also employ and compare a set of state-of-the-art evaluation metrics based on Topic Coherence.
主题建模是一种无监督学习方法，可以自动发现文本文档中隐藏的主题结构。对于文本挖掘，主题建模是一种不考虑语法和词序的独立于语言的技术。除了语义和结构问题，泰语通常被认为比其他语言更复杂。由于缺乏词分隔符和过量的复合词。单词标记化产生的错误会给文本的任何后处理(如文档检索、情感分析、机器翻译等)带来严重的问题，从而降低文本应用程序的性能。尽管词序和语义之间有很强的相关性，主题建模已经被广泛报道，它可以提取潜在的信息，即。在文档中编码的潜在主题或潜在语义。虽然之前对泰语主题建模的研究较少，但大多集中在自然语言处理(Natural language Processing, NLP)的上游过程，如将一个精炼的停止词列表应用于，或在单个特定主题建模方法上添加N-gram。据我们所知，本文首次探讨了不同的主题建模方法，即潜伏狄利克雷分配(LDA)和非负矩阵分解(NMF)，并比较了它们在泰语中的一致性。我们还采用并比较了一套基于主题一致性的最先进的评估指标。
{"title":"Discover Underlying Topics in Thai News Articles: A Comparative Study of Probabilistic and Matrix Factorization Approaches","authors":"Pimpitcha Pitichotchokphokhin, Piyawat Chuangkrud, Kongkan Kalakan, B. Suntisrivaraporn, Teerapong Leelanupab, Nont Kanungsukkasem","doi":"10.1109/ecti-con49241.2020.9158065","DOIUrl":"https://doi.org/10.1109/ecti-con49241.2020.9158065","url":null,"abstract":"Topic modeling is an unsupervised learning approach, which can automatically discover the hidden thematic structure in text documents. For text mining, topic modeling is a language-independent technique that disregards grammar and word order. Apart from semantic and structural issues, Thai language is typically considered more complex than others. Due to the lack of word delimiter and a surfeit of composite words. Errors from word tokenization can create significant problems for any post processes of text, such as document retrieval, sentiment analysis, machine translation, etc., adversely decreasing the performance of text applications. Despite a strong correlation between word ordering and semantic meaning, topic modeling has been widely reported that it can extract latent information, aka. latent topic or latent semantic, encoded in documents. Although there were few previous research works on studying topic modeling in Thai language, they mostly focused on upstream processes of Natural Language Processing (NLP) in, for example, applying a refined stop-word list to, or adding N-gram on a single specific topic modeling method. To our knowledge, this paper is the first to explore different topic modeling approaches, i.e., Latent Dirichlet Allocation (LDA) and Nonnegative Metrix Factorization (NMF), in Thai Language to compare their coherence. We also employ and compare a set of state-of-the-art evaluation metrics based on Topic Coherence.","PeriodicalId":371552,"journal":{"name":"2020 17th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology (ECTI-CON)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115205975","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 : 2020-06-01DOI: 10.1109/ecti-con49241.2020.9158208
Siriphen Pungetmongkol, E. Nantajeewarawat, Pattravadee Ploykitikoon, Paramet Tanwanont
The paper aims to investigate portfolio performance of a rebalanced portfolio and a buy-and-hold portfolio for five different market trends, i.e., up, down, sideway up, sideway down, and sideway market trends. Three asset classes, i.e., equity, bond, and gold, are considered. The portfolio initial weights for assets are determined by using the Markowitz method. Geometric Brownian motion is used for simulating future prices of each asset. Linear regression is performed to classify the simulated prices into different market trends. The rebalancing technique is applied to each trend. The risk and return performance of a rebalanced portfolio and those of a buy-and-hold portfolio are compared for each market trend.
{"title":"Portfolio Management in Different Market Trends for Thai Mutual Funds","authors":"Siriphen Pungetmongkol, E. Nantajeewarawat, Pattravadee Ploykitikoon, Paramet Tanwanont","doi":"10.1109/ecti-con49241.2020.9158208","DOIUrl":"https://doi.org/10.1109/ecti-con49241.2020.9158208","url":null,"abstract":"The paper aims to investigate portfolio performance of a rebalanced portfolio and a buy-and-hold portfolio for five different market trends, i.e., up, down, sideway up, sideway down, and sideway market trends. Three asset classes, i.e., equity, bond, and gold, are considered. The portfolio initial weights for assets are determined by using the Markowitz method. Geometric Brownian motion is used for simulating future prices of each asset. Linear regression is performed to classify the simulated prices into different market trends. The rebalancing technique is applied to each trend. The risk and return performance of a rebalanced portfolio and those of a buy-and-hold portfolio are compared for each market trend.","PeriodicalId":371552,"journal":{"name":"2020 17th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology (ECTI-CON)","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123631894","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：