Pub Date : 2022-07-06DOI: 10.1109/ICoDSA55874.2022.9862821
Nadia Dwi Puji Lestari, Esmeralda Contessa Djamal
Rainfall is influenced by climate factors such as air temperature, humidity, rainfall, wind speed, and the Southern Oscillation Index (SOI). Microclimate allows local rain to occur, so it is necessary to consider climatic variables from some observation stations. This research involved multi variables of three stations for spatial analysis. Each variable is recorded in time series. So, this paper proposed spatial and temporal analysis in predicting weekly rainfall. Spatial information was obtained from climate variables of three adjacent Meteorological, Climatology, and Geophysics Agency (BMKG) stations: Tangerang Geophysics Station, Budiarto Meteorology Station, and South Tangerang Geophysics station, for twelve years (2010-2021). The 2D Convolutional Neural Network (CNN) and Recurrent Neural Network (RNN) methods were proposed to extract spatial-temporal features from climate data. As a result, the proposed model had the best accuracy of 87.80% compared to the 1D CNN model, with an average accuracy of 80.21%. This study shows that spatial features are essential to increase accuracy because the surrounding weather variables influence each other, and there needs to be a correlation in modeling. In addition, this research also compares the proposed model with the 3D CNN method. As a result, the accuracy of the 2D CNN-RNN model outperformed the 3D CNN by 12.46% higher because 3D CNN extraction was too dependent on the extraction of spatial features and lacked optimizing temporal information.
{"title":"Rainfall Prediction using Spatial Convolutional Neural Networks and Recurrent Neural Networks","authors":"Nadia Dwi Puji Lestari, Esmeralda Contessa Djamal","doi":"10.1109/ICoDSA55874.2022.9862821","DOIUrl":"https://doi.org/10.1109/ICoDSA55874.2022.9862821","url":null,"abstract":"Rainfall is influenced by climate factors such as air temperature, humidity, rainfall, wind speed, and the Southern Oscillation Index (SOI). Microclimate allows local rain to occur, so it is necessary to consider climatic variables from some observation stations. This research involved multi variables of three stations for spatial analysis. Each variable is recorded in time series. So, this paper proposed spatial and temporal analysis in predicting weekly rainfall. Spatial information was obtained from climate variables of three adjacent Meteorological, Climatology, and Geophysics Agency (BMKG) stations: Tangerang Geophysics Station, Budiarto Meteorology Station, and South Tangerang Geophysics station, for twelve years (2010-2021). The 2D Convolutional Neural Network (CNN) and Recurrent Neural Network (RNN) methods were proposed to extract spatial-temporal features from climate data. As a result, the proposed model had the best accuracy of 87.80% compared to the 1D CNN model, with an average accuracy of 80.21%. This study shows that spatial features are essential to increase accuracy because the surrounding weather variables influence each other, and there needs to be a correlation in modeling. In addition, this research also compares the proposed model with the 3D CNN method. As a result, the accuracy of the 2D CNN-RNN model outperformed the 3D CNN by 12.46% higher because 3D CNN extraction was too dependent on the extraction of spatial features and lacked optimizing temporal information.","PeriodicalId":339135,"journal":{"name":"2022 International Conference on Data Science and Its Applications (ICoDSA)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116924574","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 : 2022-07-06DOI: 10.1109/ICoDSA55874.2022.9862830
M. V. Caya, Arthur Reimus D. Lechoncito, Gabriel Q. Deveraturda
The main goal of this study is to create a tongue print biometric system that utilizes the Oriented FAST and Rotated BRIEF (ORB) algorithm for feature extraction. The tongue print biometric system utilizes a Raspberry Pi as the microcontroller and acquires the image of tongue prints using a Raspberry Pi Camera with a Sony IMX219 8-megapixel sensor. The system initially captures the user’s tongue’s image and then uses the Contrast Limited Adaptive Histogram Equalization (CLAHE) for image pre-processing. Afterward, the ORB algorithm is used to extract the features on the Region of Interest, and then it computes the image descriptors. The descriptors are then stored in a database along with the user’s information. The data collection included thirty (30) authentic test subjects, where twenty (20) tongue prints were collected from the authentic users to train the prototype. After training, the system was tested five times on every authentic and impostor user, where the determined overall accuracy was 90.33%. Also, during the test on authentic users, the determined overall average recognition time speed of the tongue print biometric was 10.087 and the determined overall average recognition time speed when the biometric system was tested on an impostor was 10.1551 seconds. The integration of FAST and rBRIEF to ORB allowed the feature extraction algorithm to extract plenty of distributed feature points and load them fast, which led to the satisfactory accuracy rate and recognition time speeds of the tongue print biometric system.
{"title":"Recognition of Tongue Print Biometric using Oriented FAST and Rotated BRIEF (ORB)","authors":"M. V. Caya, Arthur Reimus D. Lechoncito, Gabriel Q. Deveraturda","doi":"10.1109/ICoDSA55874.2022.9862830","DOIUrl":"https://doi.org/10.1109/ICoDSA55874.2022.9862830","url":null,"abstract":"The main goal of this study is to create a tongue print biometric system that utilizes the Oriented FAST and Rotated BRIEF (ORB) algorithm for feature extraction. The tongue print biometric system utilizes a Raspberry Pi as the microcontroller and acquires the image of tongue prints using a Raspberry Pi Camera with a Sony IMX219 8-megapixel sensor. The system initially captures the user’s tongue’s image and then uses the Contrast Limited Adaptive Histogram Equalization (CLAHE) for image pre-processing. Afterward, the ORB algorithm is used to extract the features on the Region of Interest, and then it computes the image descriptors. The descriptors are then stored in a database along with the user’s information. The data collection included thirty (30) authentic test subjects, where twenty (20) tongue prints were collected from the authentic users to train the prototype. After training, the system was tested five times on every authentic and impostor user, where the determined overall accuracy was 90.33%. Also, during the test on authentic users, the determined overall average recognition time speed of the tongue print biometric was 10.087 and the determined overall average recognition time speed when the biometric system was tested on an impostor was 10.1551 seconds. The integration of FAST and rBRIEF to ORB allowed the feature extraction algorithm to extract plenty of distributed feature points and load them fast, which led to the satisfactory accuracy rate and recognition time speeds of the tongue print biometric system.","PeriodicalId":339135,"journal":{"name":"2022 International Conference on Data Science and Its Applications (ICoDSA)","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117236930","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 : 2022-07-06DOI: 10.1109/ICoDSA55874.2022.9862890
Brian Pamukti, Mutiara Faradina, Arfianto Fahmi, Nachwan Mufti Ardiansyah
For massive communication, Non orthogonal Multiple Access is considered to be implemented in next generation of 6G. Beside using radio frequencies as physical layer for wireless communication, optical wireless communication has advantages for support high speed and low error for communication. But, to server massive user, we have struggles with limited resource of power, frequency and time. In this study, we proposed the T-Fold Irregular Repetition Slotted ALOHA (IRSA) method for the Visible Light Communication (VLC) system to improve its performance. We used IRSA based on Coded Slotted ALOHA (CSA) scheme that utilities interference among users. To prove our method, we used simulation in a closed room with a room size of 6 x 6 x 6 m, using the LOS channel and transmitted power of Light Emitting Diode (LED) is 1 Watt. There are 11, 13, and 15 users with random positions that send packets in 100 time-slots based on their degree distribution. The results showed that the larger the degree distributions impact the performance. We also show that between the number of user 11 and 15, the throughput increased up to 20%, while the Packet Loss Ratio (PLR) decreased for 15 users. In addition, 6-degree distribution is used for optimal performance in 15 users.
对于大规模通信，非正交多址被认为是下一代6G的实现方式。光无线通信除了采用射频作为物理层进行无线通信外，还具有支持通信速度快、误差小的优点。但是，为了服务庞大的用户，我们不得不与有限的电力、频率和时间资源作斗争。为了提高可见光通信(VLC)系统的性能，我们提出了T-Fold不规则重复开槽ALOHA (IRSA)方法。我们采用基于编码槽ALOHA (CSA)的IRSA方案，利用用户之间的干扰。为了证明我们的方法，我们在一个房间大小为6 x 6 x 6 m的封闭房间中进行了模拟，使用LOS通道，发光二极管(LED)的发射功率为1瓦。有11、13和15个位置随机的用户，根据他们的度分布在100个时隙中发送数据包。结果表明，度分布对性能的影响越大。我们还表明，在用户数量为11和15之间，吞吐量增加到20%，而15个用户的丢包率(PLR)下降。此外，为了在15个用户中实现最佳性能，还使用了6度分布。
{"title":"Enhancement of Successive Interference Cancellation in Visible Light Communication System","authors":"Brian Pamukti, Mutiara Faradina, Arfianto Fahmi, Nachwan Mufti Ardiansyah","doi":"10.1109/ICoDSA55874.2022.9862890","DOIUrl":"https://doi.org/10.1109/ICoDSA55874.2022.9862890","url":null,"abstract":"For massive communication, Non orthogonal Multiple Access is considered to be implemented in next generation of 6G. Beside using radio frequencies as physical layer for wireless communication, optical wireless communication has advantages for support high speed and low error for communication. But, to server massive user, we have struggles with limited resource of power, frequency and time. In this study, we proposed the T-Fold Irregular Repetition Slotted ALOHA (IRSA) method for the Visible Light Communication (VLC) system to improve its performance. We used IRSA based on Coded Slotted ALOHA (CSA) scheme that utilities interference among users. To prove our method, we used simulation in a closed room with a room size of 6 x 6 x 6 m, using the LOS channel and transmitted power of Light Emitting Diode (LED) is 1 Watt. There are 11, 13, and 15 users with random positions that send packets in 100 time-slots based on their degree distribution. The results showed that the larger the degree distributions impact the performance. We also show that between the number of user 11 and 15, the throughput increased up to 20%, while the Packet Loss Ratio (PLR) decreased for 15 users. In addition, 6-degree distribution is used for optimal performance in 15 users.","PeriodicalId":339135,"journal":{"name":"2022 International Conference on Data Science and Its Applications (ICoDSA)","volume":"347 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116447895","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 : 2022-07-06DOI: 10.1109/ICoDSA55874.2022.9862918
Aldrin Jozefan Parsaoran, Satria Mandala, M. Pramudyo
Recently, Photoplethysmograph (PPG) signal has been widely considered for detecting heart-related diseases. It is because the operational cost of using this signal is relatively lower than other signals, such as the electrocardiogram (ECG). However, PPG signal is very susceptible to noise. Therefore, removing noise from the PPG signal data is a must. In most cases, the noise in this signal is much worse than the ECG signal. In addition, most existing research on denoising algorithms based on PPG signals is incomprehensive due to focusing on single denoising algorithm. This research provides a solution to the problems by proposing a performance study of three denoising algorithms for PPG signals, i.e., Savitzky Golay, Butterworth, and Finite Impulse Response (FIR). Method used to achieve the objective are literature study on denoising algorithms, conduct experiments on the proposed algorithms, measure and analyze the performance of the denoising algorithms based on three metrics, namely Signal to Noise Ratio (SNR), Peak Signal to Noise Ratio (PSNR) and Mean Square Error (MSE). Rigorous experiments have been carried out, and it is proven that Savitzky's algorithm is better than the other two algorithms (i.e., Butterworth and FIR). Savitzky has SNR:17.5 dB, PSNR: 16.80 dB and MSE: 0.19. Meanwhile, Butterworth's performance is SNR: 10.168 dB, PSNR: 9.1 dB, and MSE: 0.3. Finally, the FIR algorithm has SNR: 4.796, PSNR: 16.7, and MSE: 0.2.
{"title":"Study of Denoising Algorithms on Photoplethysmograph (PPG) Signals","authors":"Aldrin Jozefan Parsaoran, Satria Mandala, M. Pramudyo","doi":"10.1109/ICoDSA55874.2022.9862918","DOIUrl":"https://doi.org/10.1109/ICoDSA55874.2022.9862918","url":null,"abstract":"Recently, Photoplethysmograph (PPG) signal has been widely considered for detecting heart-related diseases. It is because the operational cost of using this signal is relatively lower than other signals, such as the electrocardiogram (ECG). However, PPG signal is very susceptible to noise. Therefore, removing noise from the PPG signal data is a must. In most cases, the noise in this signal is much worse than the ECG signal. In addition, most existing research on denoising algorithms based on PPG signals is incomprehensive due to focusing on single denoising algorithm. This research provides a solution to the problems by proposing a performance study of three denoising algorithms for PPG signals, i.e., Savitzky Golay, Butterworth, and Finite Impulse Response (FIR). Method used to achieve the objective are literature study on denoising algorithms, conduct experiments on the proposed algorithms, measure and analyze the performance of the denoising algorithms based on three metrics, namely Signal to Noise Ratio (SNR), Peak Signal to Noise Ratio (PSNR) and Mean Square Error (MSE). Rigorous experiments have been carried out, and it is proven that Savitzky's algorithm is better than the other two algorithms (i.e., Butterworth and FIR). Savitzky has SNR:17.5 dB, PSNR: 16.80 dB and MSE: 0.19. Meanwhile, Butterworth's performance is SNR: 10.168 dB, PSNR: 9.1 dB, and MSE: 0.3. Finally, the FIR algorithm has SNR: 4.796, PSNR: 16.7, and MSE: 0.2.","PeriodicalId":339135,"journal":{"name":"2022 International Conference on Data Science and Its Applications (ICoDSA)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128609176","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 : 2022-07-06DOI: 10.1109/ICoDSA55874.2022.9862535
Dafid, Ermatita
The need for accuracy in a prediction is a non-negotiable thing. One of the determinants of the accuracy of a prediction model is the classification method. Data mining offers various classification methods for predicting. Therefore, determining appropriate classification methods that produce high accuracy prediction model is a must. Several previous studies have shown excellent results based on influential factors for predicting students’ academic success. However, the research only focuses on one influential factor category rather than a combination of multiple influential factor categories. It becomes a serious issue since there are influential factors on the dataset that not only have one influential factor category but mostly multiple factor categories. Therefore, the best classification method for a multiple influential factor category has not been known yet. This research analyzes the performance of classification methods based on multiple categories of influential factors. The result will help the researcher find the best combination of factor category and classification method should they used. Among multiple factor category and classification methods have been tested show combination of certain classification method give the best result for certain multiple factor category.
{"title":"Determining Appropriate Classification Method Based on Influential Factors for Predicting Students’Academic Success","authors":"Dafid, Ermatita","doi":"10.1109/ICoDSA55874.2022.9862535","DOIUrl":"https://doi.org/10.1109/ICoDSA55874.2022.9862535","url":null,"abstract":"The need for accuracy in a prediction is a non-negotiable thing. One of the determinants of the accuracy of a prediction model is the classification method. Data mining offers various classification methods for predicting. Therefore, determining appropriate classification methods that produce high accuracy prediction model is a must. Several previous studies have shown excellent results based on influential factors for predicting students’ academic success. However, the research only focuses on one influential factor category rather than a combination of multiple influential factor categories. It becomes a serious issue since there are influential factors on the dataset that not only have one influential factor category but mostly multiple factor categories. Therefore, the best classification method for a multiple influential factor category has not been known yet. This research analyzes the performance of classification methods based on multiple categories of influential factors. The result will help the researcher find the best combination of factor category and classification method should they used. Among multiple factor category and classification methods have been tested show combination of certain classification method give the best result for certain multiple factor category.","PeriodicalId":339135,"journal":{"name":"2022 International Conference on Data Science and Its Applications (ICoDSA)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130632221","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 : 2022-07-06DOI: 10.1109/icodsa55874.2022.9862904
{"title":"ICoDSA 2022 Program","authors":"","doi":"10.1109/icodsa55874.2022.9862904","DOIUrl":"https://doi.org/10.1109/icodsa55874.2022.9862904","url":null,"abstract":"","PeriodicalId":339135,"journal":{"name":"2022 International Conference on Data Science and Its Applications (ICoDSA)","volume":"106 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123980894","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 : 2022-07-06DOI: 10.1109/ICoDSA55874.2022.9862828
B. I. Nasution, N. Kurniawan, S. K. Ragamustari
The COVID-19 pandemic has reached its 20th month in Indonesia and still damaged various sectors, particularly economy. The policies imposed by the government impacted mainly the stock price. exchange rate, and people mobility in Indonesia. However, there are limited studies that incorporate these variables in Indonesia context. Thus, this study investigates the relationship between the COVID-19 pandemic, stock price, exchange rate, and workplace mobility simultaneously. This study employs Vector Autoregressive (VAR) as the analysis considering its advantages in finding the causal relationship between variables and periodic interpretation using Impulse Response Function (IRF). The VAR results show that from the Granger Causality Test, it turns out that the shocks from COVID-19 positivity rate and mobility in workplaces caused the changes in stock price and exchange rate. On the other hand, the IRF results exhibit the depreciating responses of stock price and exchange rate due to the shocks of COVID-19 positivity rate and mobility are enormous in the short term. In the longer term, the stock price response needs a longer time to return to the initial condition than the exchange rate. Therefore, further policy evaluation and formulation become essential to maintain the stock price and exchange rate, mainly due to the effect of COVID-19 and workplace mobility.
{"title":"The Effects of COVID-19 and Workplace Mobility to Stock Price and Exchange Rate in Indonesia: An Econometric Approach","authors":"B. I. Nasution, N. Kurniawan, S. K. Ragamustari","doi":"10.1109/ICoDSA55874.2022.9862828","DOIUrl":"https://doi.org/10.1109/ICoDSA55874.2022.9862828","url":null,"abstract":"The COVID-19 pandemic has reached its 20th month in Indonesia and still damaged various sectors, particularly economy. The policies imposed by the government impacted mainly the stock price. exchange rate, and people mobility in Indonesia. However, there are limited studies that incorporate these variables in Indonesia context. Thus, this study investigates the relationship between the COVID-19 pandemic, stock price, exchange rate, and workplace mobility simultaneously. This study employs Vector Autoregressive (VAR) as the analysis considering its advantages in finding the causal relationship between variables and periodic interpretation using Impulse Response Function (IRF). The VAR results show that from the Granger Causality Test, it turns out that the shocks from COVID-19 positivity rate and mobility in workplaces caused the changes in stock price and exchange rate. On the other hand, the IRF results exhibit the depreciating responses of stock price and exchange rate due to the shocks of COVID-19 positivity rate and mobility are enormous in the short term. In the longer term, the stock price response needs a longer time to return to the initial condition than the exchange rate. Therefore, further policy evaluation and formulation become essential to maintain the stock price and exchange rate, mainly due to the effect of COVID-19 and workplace mobility.","PeriodicalId":339135,"journal":{"name":"2022 International Conference on Data Science and Its Applications (ICoDSA)","volume":"412 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132343045","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 : 2022-07-06DOI: 10.1109/ICoDSA55874.2022.9862903
Zayyana Nurthohari, D. I. Sensuse, Sofian Lusa
Jakarta Bus Rapid Transportation is state-own company which have services in public transportation. On October 2021, Jakarta Bus Rapid Transportation was recently trending on Twitter. Twitter public views might be utilized for the company as a decision support system for enhance and evaluate the services of the company. A sentiment analysis method may be used to examine public opinion especialy users of Jakarta Bus Rapid Transportation on Twitter. The goal of this research is to better understand Jakarta's public opinion trends about services. The researchers manually classified tweets from the Tweepy collection as Informasi, Apresiasi, Saran, or Komplain. Professionals will classify the sentiment as favorable, negative, or neutral. The data was then pre-processed to eliminate duplicates and extraneous information. The sentiment of fresh data will then be predicted using machine learning. The machine learning algorithms were then examined using a number of tests to discover which kernels and features provided the best accuracy. The result of this method shows of 92.00 percent of accuracy, 91.00 percent of precision, 92.00 percent of recall, and 2123 of support. The majority of Jakartans, according to the data, have an unfavorable impression of bus rapid transit. The majority of customers were disappointed with the services.
{"title":"Sentiment Analysis of Jakarta Bus Rapid Transportation Services using Support Vector Machine","authors":"Zayyana Nurthohari, D. I. Sensuse, Sofian Lusa","doi":"10.1109/ICoDSA55874.2022.9862903","DOIUrl":"https://doi.org/10.1109/ICoDSA55874.2022.9862903","url":null,"abstract":"Jakarta Bus Rapid Transportation is state-own company which have services in public transportation. On October 2021, Jakarta Bus Rapid Transportation was recently trending on Twitter. Twitter public views might be utilized for the company as a decision support system for enhance and evaluate the services of the company. A sentiment analysis method may be used to examine public opinion especialy users of Jakarta Bus Rapid Transportation on Twitter. The goal of this research is to better understand Jakarta's public opinion trends about services. The researchers manually classified tweets from the Tweepy collection as Informasi, Apresiasi, Saran, or Komplain. Professionals will classify the sentiment as favorable, negative, or neutral. The data was then pre-processed to eliminate duplicates and extraneous information. The sentiment of fresh data will then be predicted using machine learning. The machine learning algorithms were then examined using a number of tests to discover which kernels and features provided the best accuracy. The result of this method shows of 92.00 percent of accuracy, 91.00 percent of precision, 92.00 percent of recall, and 2123 of support. The majority of Jakartans, according to the data, have an unfavorable impression of bus rapid transit. The majority of customers were disappointed with the services.","PeriodicalId":339135,"journal":{"name":"2022 International Conference on Data Science and Its Applications (ICoDSA)","volume":"385 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123515928","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 : 2022-07-06DOI: 10.1109/ICoDSA55874.2022.9862862
Jaka Pratama, A. Choiruddin
Spatial point pattern is randomly arranged collection of points distributed over space, such as the locations of a tree species in a forest. Such a study is also commonly known as Species Distribution Modeling (SDM), where the main concern is to relate the distribution of tree species and environmental variables. Within spatial point process framework, SDM is closely related to modeling the intensity of spatial point process. The standard technique for parameter estimation of the intensity is by method of Maximum Likelihood Estimation (MLE) employing Berman-Turner Approximation, resulting in Poisson-based regression. However, this technique could raise an issue due to a large number of dummy points required in the approximation since large number of dummy points relates to excessive zeroes in response variable. Previous studies suggest the application of Zero Inflated Poisson (ZIP) regression over Poisson regression to model response variable with excessive zeroes. This study compares Poisson and ZIP-based method for modelling the distribution of Beilschmiedia Pendula tree with respect to environmental covariates. We compared both techniques by Bayesian Information Criteria (BIC) and concluded that the ZIP-based method performs better mainly due to excessive zeroes from dummy points. In addition, elevation and gradient affect significantly the distribution of Beilschmiedia Pendula tree.
{"title":"Species Distribution Modeling with Spatial Point Process: Comparing Poisson and Zero Inflated Poisson-Based Algorithms","authors":"Jaka Pratama, A. Choiruddin","doi":"10.1109/ICoDSA55874.2022.9862862","DOIUrl":"https://doi.org/10.1109/ICoDSA55874.2022.9862862","url":null,"abstract":"Spatial point pattern is randomly arranged collection of points distributed over space, such as the locations of a tree species in a forest. Such a study is also commonly known as Species Distribution Modeling (SDM), where the main concern is to relate the distribution of tree species and environmental variables. Within spatial point process framework, SDM is closely related to modeling the intensity of spatial point process. The standard technique for parameter estimation of the intensity is by method of Maximum Likelihood Estimation (MLE) employing Berman-Turner Approximation, resulting in Poisson-based regression. However, this technique could raise an issue due to a large number of dummy points required in the approximation since large number of dummy points relates to excessive zeroes in response variable. Previous studies suggest the application of Zero Inflated Poisson (ZIP) regression over Poisson regression to model response variable with excessive zeroes. This study compares Poisson and ZIP-based method for modelling the distribution of Beilschmiedia Pendula tree with respect to environmental covariates. We compared both techniques by Bayesian Information Criteria (BIC) and concluded that the ZIP-based method performs better mainly due to excessive zeroes from dummy points. In addition, elevation and gradient affect significantly the distribution of Beilschmiedia Pendula tree.","PeriodicalId":339135,"journal":{"name":"2022 International Conference on Data Science and Its Applications (ICoDSA)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124030719","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 : 2022-07-06DOI: 10.1109/ICoDSA55874.2022.9862858
R. Chitty, Keerthi Gunawikrama, Harinda Fernando
Finance Companies (FC’s), play a pivotal role in the economy of Sri Lanka, by serving the under banked and non-banked segments of the society. The business model entails lending to the bottom of the pyramid, that leads to the acceptance of higher credit risk at a higher yield that inevitably leads to lower asset quality. The focus on this customer segment has lead to an increase in non performing loans among FCs in the recent past. Due to several challenges facing the industry, including intense competition and lack of experienced credit officers, the FC’s have been seeking options to automate evaluation of credit worthiness at the point of loan origination. This work is an attempt to develop a machine learning based loan default prediction system to improve credit decisions. Several traditional machine learning algorithms are chosen, trained and validated by using real world data set related to vehicle leasing, obtained from one of the leading FCs in Sri Lanka. The data set consists of 100,000 cases having 29 attributes each. Models are compared for accuracy, sensitivity, specificity and robustness. The model using Support Vector Machine and Random Forest produces comparatively promising results. Further work is recommended to generalize the model for economic cycles and shocks using micro and macro economic variables.
{"title":"Development of Loan Default Prediction Model for Finance Companies in Sri Lanka – A Case Study","authors":"R. Chitty, Keerthi Gunawikrama, Harinda Fernando","doi":"10.1109/ICoDSA55874.2022.9862858","DOIUrl":"https://doi.org/10.1109/ICoDSA55874.2022.9862858","url":null,"abstract":"Finance Companies (FC’s), play a pivotal role in the economy of Sri Lanka, by serving the under banked and non-banked segments of the society. The business model entails lending to the bottom of the pyramid, that leads to the acceptance of higher credit risk at a higher yield that inevitably leads to lower asset quality. The focus on this customer segment has lead to an increase in non performing loans among FCs in the recent past. Due to several challenges facing the industry, including intense competition and lack of experienced credit officers, the FC’s have been seeking options to automate evaluation of credit worthiness at the point of loan origination. This work is an attempt to develop a machine learning based loan default prediction system to improve credit decisions. Several traditional machine learning algorithms are chosen, trained and validated by using real world data set related to vehicle leasing, obtained from one of the leading FCs in Sri Lanka. The data set consists of 100,000 cases having 29 attributes each. Models are compared for accuracy, sensitivity, specificity and robustness. The model using Support Vector Machine and Random Forest produces comparatively promising results. Further work is recommended to generalize the model for economic cycles and shocks using micro and macro economic variables.","PeriodicalId":339135,"journal":{"name":"2022 International Conference on Data Science and Its Applications (ICoDSA)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126130953","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：