Pub Date : 2023-02-16DOI: 10.1109/ICCoSITE57641.2023.10127770
B. Wijanarko, Dina Fitria Murad, Y. Heryadi, C. Tho, Kiyota Hashimoto
Automated language translation involving low-resource language has gained wide interest from many research communities in the past decade. One lesson learned from the past COVID-19 pandemic, particularly in Indonesia, is that many local Governments have to release regular public announcements to keep people following health protocol especially when they are in public areas. Many studies showed some evidence that rural people in Indonesia which covers a large proportion of Indonesia’s population, feel more convenience receiving official announcements in their local language. However, translating official announcement from the national language to many local languages in Indonesia require many experienced bilingual translators and time. This paper presents exploration results in developing an automated language translator model to translate texts in Bahasa Indonesia to the Sundanese language. In particular, this study aims to explore the effect of ReLU, Sigmoid, and Tanh activation functions of the Vanilla Transformer Model on its translation performance. The experiment results showed that the activation function under study gives similar training accuracy (0.98). However, ReLU achieves better performance than Tanh in terms of validation accuracy, training loss, and validation loss.
{"title":"Exploring the Effect of Activation Function on Transformer Model Performance for Official Announcement Translator from Indonesian to Sundanese Languages","authors":"B. Wijanarko, Dina Fitria Murad, Y. Heryadi, C. Tho, Kiyota Hashimoto","doi":"10.1109/ICCoSITE57641.2023.10127770","DOIUrl":"https://doi.org/10.1109/ICCoSITE57641.2023.10127770","url":null,"abstract":"Automated language translation involving low-resource language has gained wide interest from many research communities in the past decade. One lesson learned from the past COVID-19 pandemic, particularly in Indonesia, is that many local Governments have to release regular public announcements to keep people following health protocol especially when they are in public areas. Many studies showed some evidence that rural people in Indonesia which covers a large proportion of Indonesia’s population, feel more convenience receiving official announcements in their local language. However, translating official announcement from the national language to many local languages in Indonesia require many experienced bilingual translators and time. This paper presents exploration results in developing an automated language translator model to translate texts in Bahasa Indonesia to the Sundanese language. In particular, this study aims to explore the effect of ReLU, Sigmoid, and Tanh activation functions of the Vanilla Transformer Model on its translation performance. The experiment results showed that the activation function under study gives similar training accuracy (0.98). However, ReLU achieves better performance than Tanh in terms of validation accuracy, training loss, and validation loss.","PeriodicalId":256184,"journal":{"name":"2023 International Conference on Computer Science, Information Technology and Engineering (ICCoSITE)","volume":"108 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132240644","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 : 2023-02-16DOI: 10.1109/ICCoSITE57641.2023.10127689
Md Musfiq Us Saleheen, Md Rabbul Fahad, R. Khan
Robotic arms are highly effective for industries that demand quick and reliable performance. These efficient devices are essentially automated systems that, unlike humans, do not get tired or need a rest. These machines have been used for many years but have recently progressed significantly with the advancement of complex sensors. Robotic arms of today come with various sensors that let them move around and react quickly in their working areas. This paper introduces a human hand gesture-controlled automatic low-cost robotic arm. In this proposed system, an Arduino Mega microcontroller gets the information from all the sensors and correctly manages the servomotor with the help of the value of sensors. All the sensors required to control the various servos on the robotic arm are placed into a hand glove. The robotic arm is operated in this system by two flex sensors. One flex sensor is linked to the glove’s forefinger section to manage the arm’s claw, and another is attached to the middle finger section of the glove to regulate the arm’s wrist. A gyroscope is also pinned to the glove to track the movement of the forearm and base. As a result, the base servo moves clockwise or counterclockwise depending on whether the hand glove is angled right or left. However, if the hand glove is angled upward or downward, the gyroscope data will cause the forearm servo to rotate either clockwise or counterclockwise. The sensors’ values are converted to the servo motors’ rotational degrees. The sensors’ values are converted to the servo motors’ rotational degrees. The claw, wrist, forearm servos and base of the proposed robotic device can rotate up to 900, 450, 1200 and 1800 degrees, respectively.
{"title":"Gesture-Controlled Robotic Arm","authors":"Md Musfiq Us Saleheen, Md Rabbul Fahad, R. Khan","doi":"10.1109/ICCoSITE57641.2023.10127689","DOIUrl":"https://doi.org/10.1109/ICCoSITE57641.2023.10127689","url":null,"abstract":"Robotic arms are highly effective for industries that demand quick and reliable performance. These efficient devices are essentially automated systems that, unlike humans, do not get tired or need a rest. These machines have been used for many years but have recently progressed significantly with the advancement of complex sensors. Robotic arms of today come with various sensors that let them move around and react quickly in their working areas. This paper introduces a human hand gesture-controlled automatic low-cost robotic arm. In this proposed system, an Arduino Mega microcontroller gets the information from all the sensors and correctly manages the servomotor with the help of the value of sensors. All the sensors required to control the various servos on the robotic arm are placed into a hand glove. The robotic arm is operated in this system by two flex sensors. One flex sensor is linked to the glove’s forefinger section to manage the arm’s claw, and another is attached to the middle finger section of the glove to regulate the arm’s wrist. A gyroscope is also pinned to the glove to track the movement of the forearm and base. As a result, the base servo moves clockwise or counterclockwise depending on whether the hand glove is angled right or left. However, if the hand glove is angled upward or downward, the gyroscope data will cause the forearm servo to rotate either clockwise or counterclockwise. The sensors’ values are converted to the servo motors’ rotational degrees. The sensors’ values are converted to the servo motors’ rotational degrees. The claw, wrist, forearm servos and base of the proposed robotic device can rotate up to 900, 450, 1200 and 1800 degrees, respectively.","PeriodicalId":256184,"journal":{"name":"2023 International Conference on Computer Science, Information Technology and Engineering (ICCoSITE)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132988696","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 : 2023-02-16DOI: 10.1109/ICCoSITE57641.2023.10127750
Thia Anissa, H. A. Nugroho, I. Soesanti
Thyroid nodule composition is one of the determinants of thyroid cancer malignancy. Nodules in the thyroid can be detected by ultrasonography, which is one of the most sensitive imaging methods. However, imaging methods by ultrasonography are susceptible to doctors’ experiences, levels, and other factors. Therefore, a more objective diagnostic system is intended to assist doctors in creating the decision. This study developed a method to help experts define the composition characteristics. The experts have already cropped the dataset and then moved to preprocess using an adaptive median filter. Subsequently, the data were classified with the improved VGG16 into four categories those are cystic, solid, complex, and spongiform. The testing result procured 99.65% for accuracy, 99.98% for the micro area under the curve, and 99.99% for the macro area under the curve. These results indicate that our proposed method can be used in a small dataset to help doctors or experts identify the nodule’s characteristics.
{"title":"Improved VGG-16 for Classifying Thyroid Nodule on Thyroid Ultrasound Images","authors":"Thia Anissa, H. A. Nugroho, I. Soesanti","doi":"10.1109/ICCoSITE57641.2023.10127750","DOIUrl":"https://doi.org/10.1109/ICCoSITE57641.2023.10127750","url":null,"abstract":"Thyroid nodule composition is one of the determinants of thyroid cancer malignancy. Nodules in the thyroid can be detected by ultrasonography, which is one of the most sensitive imaging methods. However, imaging methods by ultrasonography are susceptible to doctors’ experiences, levels, and other factors. Therefore, a more objective diagnostic system is intended to assist doctors in creating the decision. This study developed a method to help experts define the composition characteristics. The experts have already cropped the dataset and then moved to preprocess using an adaptive median filter. Subsequently, the data were classified with the improved VGG16 into four categories those are cystic, solid, complex, and spongiform. The testing result procured 99.65% for accuracy, 99.98% for the micro area under the curve, and 99.99% for the macro area under the curve. These results indicate that our proposed method can be used in a small dataset to help doctors or experts identify the nodule’s characteristics.","PeriodicalId":256184,"journal":{"name":"2023 International Conference on Computer Science, Information Technology and Engineering (ICCoSITE)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127660423","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 : 2023-02-16DOI: 10.1109/ICCoSITE57641.2023.10127741
Alfarizi Wiranata, N. Karna, A. Irawan, Ian Agung Prakoso
Cyber-attack is an inevitable risk from any business in internet era. Cyber-attack can originate from both internal employees and external sources. The office network can be in danger due to cyber-attack and can disturb the workflow. This issue can be prevented by Host Intrusion Prevention System (HIPS) Snort that secures the network through smart security in a box system with Raspberry Pi as the firewall to protect the user devices against Denial of Service (DoS) attacks. Raspberry Pi is configured to be the firewall with installing the HIPS Snort as a defence system to protect the user's work devices. This smart security in a box is installed between the user's devices before connected to the office network. The smart security in a box detects the misuse in the network for all data packets that are suspected of being DoS attacks and drops them. DoS attacks using SYN Flood and UDP Flood are going to put Snort to the test. The successful client connection when Snort is running are only the average of 48.60% and 46.31% for SYN Flood and UDP Flood attack respectively. When Snort is running, HIPS Snort can drop the average of 41.48% of SYN Flood attack and 28.27% of UDP Flood attack packets. CPU and Memory usage are higher when Snort is running. DoS SYN Flood attack consumes more CPU and Memory usage of Raspberry Pi with the average of 83.60% and 76.75% respectively when Snort is running.
{"title":"Implementation and Analysis of Network Security in Raspberry Pi against DOS Attack with HIPS Snort","authors":"Alfarizi Wiranata, N. Karna, A. Irawan, Ian Agung Prakoso","doi":"10.1109/ICCoSITE57641.2023.10127741","DOIUrl":"https://doi.org/10.1109/ICCoSITE57641.2023.10127741","url":null,"abstract":"Cyber-attack is an inevitable risk from any business in internet era. Cyber-attack can originate from both internal employees and external sources. The office network can be in danger due to cyber-attack and can disturb the workflow. This issue can be prevented by Host Intrusion Prevention System (HIPS) Snort that secures the network through smart security in a box system with Raspberry Pi as the firewall to protect the user devices against Denial of Service (DoS) attacks. Raspberry Pi is configured to be the firewall with installing the HIPS Snort as a defence system to protect the user's work devices. This smart security in a box is installed between the user's devices before connected to the office network. The smart security in a box detects the misuse in the network for all data packets that are suspected of being DoS attacks and drops them. DoS attacks using SYN Flood and UDP Flood are going to put Snort to the test. The successful client connection when Snort is running are only the average of 48.60% and 46.31% for SYN Flood and UDP Flood attack respectively. When Snort is running, HIPS Snort can drop the average of 41.48% of SYN Flood attack and 28.27% of UDP Flood attack packets. CPU and Memory usage are higher when Snort is running. DoS SYN Flood attack consumes more CPU and Memory usage of Raspberry Pi with the average of 83.60% and 76.75% respectively when Snort is running.","PeriodicalId":256184,"journal":{"name":"2023 International Conference on Computer Science, Information Technology and Engineering (ICCoSITE)","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121478311","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 : 2023-02-16DOI: 10.1109/ICCoSITE57641.2023.10127712
Imelda Uli Vistalina Simanjuntak, T. Yunita, Chairunnisa Chairunnisa, A. Kurniawan
One of the methods used to increase channel capacity is the Non-Orthogonal Multiple Access (NOMA) methods. The increase in channel capacity in PD-NOMA multiple access means that the maximum power used by each user in each narrow band achieves increased power efficiency. Therefore, this research focuses on facilitating the increase in channel capacity by designing an antenna that produces a radiation pattern with many sharp main lobes and high gain. The multi-beam radiation pattern is obtained by using edge weighting. Therefore the antenna is designed with odd and even numbers represented by 1x5 and 1x6 antennas on the microstrip patch rectangular antenna at a frequency of 2.6 GHz. The effect of parasitic and non-parasite effects on the radiation pattern of a planar antenna with edge modeling is manageable in the main lobe, only in the side level lobe, which is not too significant. However, the number of main lobes on the 1x6 antenna has seven beam patterns, and the 1x5 antenna has five. The highest gain value obtains in the 1x6 design with a parasitic of 6.01 dB. This value meets the Gain requirements for 5G cellular systems.
{"title":"Radiation Pattern Characteristics Of 1x5 And 1x6 Planar Microstrip Antennas With Edge Weighting At A Frequency Of 2.6 Ghz","authors":"Imelda Uli Vistalina Simanjuntak, T. Yunita, Chairunnisa Chairunnisa, A. Kurniawan","doi":"10.1109/ICCoSITE57641.2023.10127712","DOIUrl":"https://doi.org/10.1109/ICCoSITE57641.2023.10127712","url":null,"abstract":"One of the methods used to increase channel capacity is the Non-Orthogonal Multiple Access (NOMA) methods. The increase in channel capacity in PD-NOMA multiple access means that the maximum power used by each user in each narrow band achieves increased power efficiency. Therefore, this research focuses on facilitating the increase in channel capacity by designing an antenna that produces a radiation pattern with many sharp main lobes and high gain. The multi-beam radiation pattern is obtained by using edge weighting. Therefore the antenna is designed with odd and even numbers represented by 1x5 and 1x6 antennas on the microstrip patch rectangular antenna at a frequency of 2.6 GHz. The effect of parasitic and non-parasite effects on the radiation pattern of a planar antenna with edge modeling is manageable in the main lobe, only in the side level lobe, which is not too significant. However, the number of main lobes on the 1x6 antenna has seven beam patterns, and the 1x5 antenna has five. The highest gain value obtains in the 1x6 design with a parasitic of 6.01 dB. This value meets the Gain requirements for 5G cellular systems.","PeriodicalId":256184,"journal":{"name":"2023 International Conference on Computer Science, Information Technology and Engineering (ICCoSITE)","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115765620","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 : 2023-02-16DOI: 10.1109/ICCoSITE57641.2023.10127721
Suriansyah B, A. A. Ilham, A. W. Paundu
Data growth is increasing day by day, so the data stored in the data warehouse is increasingly piling up. When data is displayed on the dashboard or information system, performance is slow because the process of loading queries from the data warehouse to the information system will access all the data stored in the data warehouse tables. This causes the speed of loading data on information systems to decrease, so optimization is needed in the data warehouse so that the load process becomes lighter even though data growth is increasing. In this research, a scheduling algorithm will be created in Hadoop whose job is to execute the transform extraction process and load summary data into several tables. Aims to streamline and optimized the Extract, Transform, Load (ETL) process to the data warehouse and reduce the volume of data in one table, then will be indexed according to the primary key in each table so that when data is joined to several tables it can be executed quickly. After testing by querying data with the same goal but different tables, namely tables that are optimized and unoptimized produce a query time of 1.418 seconds, while tables unoptimized have a query time of 2.418 seconds. Well as testing the speed of loading data into the information system by comparing the throughput of systems that are optimized and those that are unoptimized have an average throughput difference of 85%. With these results, it can be concluded that the speed in loading data into the information system has been successfully optimized by looking at this comparison.
{"title":"Optimization of Data Warehouse Architecture to Improve Information System Performance","authors":"Suriansyah B, A. A. Ilham, A. W. Paundu","doi":"10.1109/ICCoSITE57641.2023.10127721","DOIUrl":"https://doi.org/10.1109/ICCoSITE57641.2023.10127721","url":null,"abstract":"Data growth is increasing day by day, so the data stored in the data warehouse is increasingly piling up. When data is displayed on the dashboard or information system, performance is slow because the process of loading queries from the data warehouse to the information system will access all the data stored in the data warehouse tables. This causes the speed of loading data on information systems to decrease, so optimization is needed in the data warehouse so that the load process becomes lighter even though data growth is increasing. In this research, a scheduling algorithm will be created in Hadoop whose job is to execute the transform extraction process and load summary data into several tables. Aims to streamline and optimized the Extract, Transform, Load (ETL) process to the data warehouse and reduce the volume of data in one table, then will be indexed according to the primary key in each table so that when data is joined to several tables it can be executed quickly. After testing by querying data with the same goal but different tables, namely tables that are optimized and unoptimized produce a query time of 1.418 seconds, while tables unoptimized have a query time of 2.418 seconds. Well as testing the speed of loading data into the information system by comparing the throughput of systems that are optimized and those that are unoptimized have an average throughput difference of 85%. With these results, it can be concluded that the speed in loading data into the information system has been successfully optimized by looking at this comparison.","PeriodicalId":256184,"journal":{"name":"2023 International Conference on Computer Science, Information Technology and Engineering (ICCoSITE)","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125979117","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 : 2023-02-16DOI: 10.1109/ICCoSITE57641.2023.10127695
Muhammad Hasan Albana, H. Guntur, A. Putra
Excessive heat will reduce the torque, power, and efficiency of the in-wheel motor, for example, the in-wheel motor used for an electric scooter. This study discusses an effort to reduce the temperature of the in-wheel motor of an electric scooter by making grooves on the cover surface. If the temperature of the in-wheel motor cover is low, it will reduce the temperature of the in-wheel motor as a whole. This research uses the finite element method: steady-state thermal analysis and transient thermal analysis. The software used is Ansys Workbench. Suppose a heat flux of 1000 W/m2 is inside the in- wheel motor, the metal used as a cover is aluminum, 22°C is the starting temperature, and the ambient temperature is 28°C. The results showed that the grooves on the in-wheel motor cover caused the temperature to be reduced by 8.4%, and the total heat flux increased by 8.3% compared to the standard in-wheel motor cover (cover without modifications). In this study, the transient thermal analysis described the heat transfer process better than a steady-state thermal analysis.
{"title":"The effect of grooves on the in-wheel motor cover of an electric scooter to increase the heat transfer process","authors":"Muhammad Hasan Albana, H. Guntur, A. Putra","doi":"10.1109/ICCoSITE57641.2023.10127695","DOIUrl":"https://doi.org/10.1109/ICCoSITE57641.2023.10127695","url":null,"abstract":"Excessive heat will reduce the torque, power, and efficiency of the in-wheel motor, for example, the in-wheel motor used for an electric scooter. This study discusses an effort to reduce the temperature of the in-wheel motor of an electric scooter by making grooves on the cover surface. If the temperature of the in-wheel motor cover is low, it will reduce the temperature of the in-wheel motor as a whole. This research uses the finite element method: steady-state thermal analysis and transient thermal analysis. The software used is Ansys Workbench. Suppose a heat flux of 1000 W/m2 is inside the in- wheel motor, the metal used as a cover is aluminum, 22°C is the starting temperature, and the ambient temperature is 28°C. The results showed that the grooves on the in-wheel motor cover caused the temperature to be reduced by 8.4%, and the total heat flux increased by 8.3% compared to the standard in-wheel motor cover (cover without modifications). In this study, the transient thermal analysis described the heat transfer process better than a steady-state thermal analysis.","PeriodicalId":256184,"journal":{"name":"2023 International Conference on Computer Science, Information Technology and Engineering (ICCoSITE)","volume":"114 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126033615","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 : 2023-02-16DOI: 10.1109/ICCoSITE57641.2023.10127749
Hazel Fauzan Delfianto, Yakob Utama Chandra
As subscription services rise as a popular choice among companies to monetize their services, they must understand the factors that contribute to the success of a subscription service. Consumer behavior and preferences will play an important role to gauge the factors which will eventually lead to a consumer’s continuous use intention of the subscription service. To better understand the customer’s behavior pattern, we built this research using a quantitative method through a questionnaire to find which factors play an important role in a successful subscription service. The study was conducted using a quantitative method through a questionnaire, with 7 hypotheses tested using a research model that expands on an existing model by fellow researchers. The data was collected from 563 valid responses, which were processed using the SmartPLS program. The program allowed for the calculation of the reliability and validity of our data, as well as the path coefficients of the data, which indicated which factors are significant. Upon running the program, we can conclude that the variables are reliable and that all the paths are shown to be significant. Lastly, the program will enable us to calculate the Q2 prediction value to measure the predictive relevance of several sub-variables. The results show that the measurement model has low predictive relevance due to its low Q2 values. The state of the art in the subject of subscription services has advanced in recent years, and a growing amount of research has been done to examine the factors that determine a subscription service's success. By highlighting the importance of perceived value in affecting consumer behavior when it comes to subscription services, this research contributes to existing literature. This study shows the need for a deeper understanding of consumer behavior in subscription services and provides a foundation for further research in this field. By exploring the impact of perceived value on consumer behavior and purchase intentions, this study contributes to a deeper understanding of the underlying factors that drive the success of a subscription service.
{"title":"The Impact of Key Factors on Subscription Service Adoption and Retention in Indonesia","authors":"Hazel Fauzan Delfianto, Yakob Utama Chandra","doi":"10.1109/ICCoSITE57641.2023.10127749","DOIUrl":"https://doi.org/10.1109/ICCoSITE57641.2023.10127749","url":null,"abstract":"As subscription services rise as a popular choice among companies to monetize their services, they must understand the factors that contribute to the success of a subscription service. Consumer behavior and preferences will play an important role to gauge the factors which will eventually lead to a consumer’s continuous use intention of the subscription service. To better understand the customer’s behavior pattern, we built this research using a quantitative method through a questionnaire to find which factors play an important role in a successful subscription service. The study was conducted using a quantitative method through a questionnaire, with 7 hypotheses tested using a research model that expands on an existing model by fellow researchers. The data was collected from 563 valid responses, which were processed using the SmartPLS program. The program allowed for the calculation of the reliability and validity of our data, as well as the path coefficients of the data, which indicated which factors are significant. Upon running the program, we can conclude that the variables are reliable and that all the paths are shown to be significant. Lastly, the program will enable us to calculate the Q2 prediction value to measure the predictive relevance of several sub-variables. The results show that the measurement model has low predictive relevance due to its low Q2 values. The state of the art in the subject of subscription services has advanced in recent years, and a growing amount of research has been done to examine the factors that determine a subscription service's success. By highlighting the importance of perceived value in affecting consumer behavior when it comes to subscription services, this research contributes to existing literature. This study shows the need for a deeper understanding of consumer behavior in subscription services and provides a foundation for further research in this field. By exploring the impact of perceived value on consumer behavior and purchase intentions, this study contributes to a deeper understanding of the underlying factors that drive the success of a subscription service.","PeriodicalId":256184,"journal":{"name":"2023 International Conference on Computer Science, Information Technology and Engineering (ICCoSITE)","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126823162","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 : 2023-02-16DOI: 10.1109/ICCoSITE57641.2023.10127739
Alexander D. Maralit, Alexel A. Imperial, Rinoa T. Cayangyang, Jose B. Tan, Roselyn A. Maaño, Rodrigo C. Belleza, P. J. D. de Castro, David Eric S. Oreta
Honeybee colonies missing their queens are more likely to swarm and experience a fall in population. Bee growers in the Philippines still utilize traditional methods to determine the health of a hive. Traditional methods lead to difficulties if a hive goes without a queen for an extended period. The study focuses on how sound data may be used as input to a CNN-based architecture to determine whether a beehive has a queen. The research involves preparing audio files for conversion into a spectrogram, converting audio data into a spectrogram, converting the spectrogram into a Mel frequency cepstral coefficient, constructing and training a model for a feature based on the features of the spectrogram that is provided, and, as the last step, assessing the model with audio files that are different from the data used in the study. The study employs four CNN-based architectures for the training and evaluating of the model containing audio recordings taken from various beehives, each of which either lacked a queen or had one present. The simplified CNN model has an accuracy of 99.88% when predicting the sound of a queen-right hive, and it has an accuracy of 99.72% when predicting the sound of a queen-less hive.
{"title":"QueenBuzz: A CNN-based architecture for Sound Processing of Queenless Beehive Towards European Apis Mellifera Bee Colonies' Survivability","authors":"Alexander D. Maralit, Alexel A. Imperial, Rinoa T. Cayangyang, Jose B. Tan, Roselyn A. Maaño, Rodrigo C. Belleza, P. J. D. de Castro, David Eric S. Oreta","doi":"10.1109/ICCoSITE57641.2023.10127739","DOIUrl":"https://doi.org/10.1109/ICCoSITE57641.2023.10127739","url":null,"abstract":"Honeybee colonies missing their queens are more likely to swarm and experience a fall in population. Bee growers in the Philippines still utilize traditional methods to determine the health of a hive. Traditional methods lead to difficulties if a hive goes without a queen for an extended period. The study focuses on how sound data may be used as input to a CNN-based architecture to determine whether a beehive has a queen. The research involves preparing audio files for conversion into a spectrogram, converting audio data into a spectrogram, converting the spectrogram into a Mel frequency cepstral coefficient, constructing and training a model for a feature based on the features of the spectrogram that is provided, and, as the last step, assessing the model with audio files that are different from the data used in the study. The study employs four CNN-based architectures for the training and evaluating of the model containing audio recordings taken from various beehives, each of which either lacked a queen or had one present. The simplified CNN model has an accuracy of 99.88% when predicting the sound of a queen-right hive, and it has an accuracy of 99.72% when predicting the sound of a queen-less hive.","PeriodicalId":256184,"journal":{"name":"2023 International Conference on Computer Science, Information Technology and Engineering (ICCoSITE)","volume":"59 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127485932","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 : 2023-02-16DOI: 10.1109/ICCoSITE57641.2023.10127842
Y. Septiana, Yoga Handoko Agustin, Muhammad Nashir Mudzakir, A. Mulyani, Dini Destiani Siti Fatimah, Indri Tri Julianto
Based on the epidemiological update or weekly spread of Covid-19 on 23 February 2021, Indonesia was ranked second in the Southeast Asia region in the highest new case reporting. The Indonesian government has taken various countermeasures to suppress the spread of Covid-19, starting from implementing health protocols for the public in public places to the Enforcement of Restrictions on Community Activities. Health facilities are shared facilities included in the essential sector, allowing them to operate 100% by regulating operating hours and capacity and implementing more stringent health protocols. This study aimed to implement a patient classification model using the C4.5 algorithm in determining emergency patients in the maternity hospital queue system. The methodology used in this study is the Cross-Industry Standard Process for Data Mining (CRISP-DM). In contrast, the evaluation of classification data uses the Confusion Matrix and Receiver Operating Characteristics (ROC). The implementation of the C4.5 algorithm in the Maternity Hospital Queue System is used to classify emergency and nonemergency patients. The classification accuracy level obtained in this study was 97.08%, and the AUC value received was 0.984.
{"title":"Implementation of Classification Algorithm C4.5 in Determining the Emergency Patient in the Maternity Hospital Queue System","authors":"Y. Septiana, Yoga Handoko Agustin, Muhammad Nashir Mudzakir, A. Mulyani, Dini Destiani Siti Fatimah, Indri Tri Julianto","doi":"10.1109/ICCoSITE57641.2023.10127842","DOIUrl":"https://doi.org/10.1109/ICCoSITE57641.2023.10127842","url":null,"abstract":"Based on the epidemiological update or weekly spread of Covid-19 on 23 February 2021, Indonesia was ranked second in the Southeast Asia region in the highest new case reporting. The Indonesian government has taken various countermeasures to suppress the spread of Covid-19, starting from implementing health protocols for the public in public places to the Enforcement of Restrictions on Community Activities. Health facilities are shared facilities included in the essential sector, allowing them to operate 100% by regulating operating hours and capacity and implementing more stringent health protocols. This study aimed to implement a patient classification model using the C4.5 algorithm in determining emergency patients in the maternity hospital queue system. The methodology used in this study is the Cross-Industry Standard Process for Data Mining (CRISP-DM). In contrast, the evaluation of classification data uses the Confusion Matrix and Receiver Operating Characteristics (ROC). The implementation of the C4.5 algorithm in the Maternity Hospital Queue System is used to classify emergency and nonemergency patients. The classification accuracy level obtained in this study was 97.08%, and the AUC value received was 0.984.","PeriodicalId":256184,"journal":{"name":"2023 International Conference on Computer Science, Information Technology and Engineering (ICCoSITE)","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124129895","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：