Pub Date : 2020-06-01DOI: 10.1109/ecti-con49241.2020.9158310
Kann Yingprayoon, Sansiri Tanachutiwat
A simple spectrometer is constructed using Light Emitting Diodes of different colors as sources for this spectrometer. In order to calibrate the light sources, different types of LEDs are connected to the circuit of constant voltage to give lights of different colors. The colors of these LEDs are Blue, Green, Yellow, Orange and red with different wavelengths. The emission spectra of all LEDs were obtained from commercial standard companies to give the peak wavelengths of each LED, 400nm (Violet), 470nm (Blue), 525nm (Green), 574nm (Yellow), 590nm (Orange), 610nm (Orange), 625nm (Red), 700nm (Dark Red) respectively. These peak wavelengths are used as reference light sources for this spectrometer. The results of Absorption spectrum measurement show similar spectrum from standard measurement. Raspberry Pi microcontroller was used in this study to measure, to analyze data and display the absorption spectrum. This low-cost spectrometer is good enough for education which can be used in the normal schools or educational institutions using LEDs of several standard peak wavelengths.
{"title":"Simple Spectrometer for Education Using Microcontroller","authors":"Kann Yingprayoon, Sansiri Tanachutiwat","doi":"10.1109/ecti-con49241.2020.9158310","DOIUrl":"https://doi.org/10.1109/ecti-con49241.2020.9158310","url":null,"abstract":"A simple spectrometer is constructed using Light Emitting Diodes of different colors as sources for this spectrometer. In order to calibrate the light sources, different types of LEDs are connected to the circuit of constant voltage to give lights of different colors. The colors of these LEDs are Blue, Green, Yellow, Orange and red with different wavelengths. The emission spectra of all LEDs were obtained from commercial standard companies to give the peak wavelengths of each LED, 400nm (Violet), 470nm (Blue), 525nm (Green), 574nm (Yellow), 590nm (Orange), 610nm (Orange), 625nm (Red), 700nm (Dark Red) respectively. These peak wavelengths are used as reference light sources for this spectrometer. The results of Absorption spectrum measurement show similar spectrum from standard measurement. Raspberry Pi microcontroller was used in this study to measure, to analyze data and display the absorption spectrum. This low-cost spectrometer is good enough for education which can be used in the normal schools or educational institutions using LEDs of several standard peak wavelengths.","PeriodicalId":371552,"journal":{"name":"2020 17th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology (ECTI-CON)","volume":"347 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":"115282049","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-con49241.2020.9158334
B. Pravalpruk, S. Watcharabutsarakham
Hand writing and hand drawing are natural ways to take note. A pen and papers are used to make a note for a long time. In digital era, the notes are often converted into a durable and formal format for further use. Therefore, the conversion application was developed in many fields with many skill such as handwritten recognition, object recognition, object classification, and others. In this paper, we demonstrate a method to classify connected components as flowchart and text. We use the Online Handwritten Flowchart Dataset (OHFD) which contained 419 handwritten flowcharts to benchmark our methodology. The result shown our classification technique get F1-score 77.6%.
{"title":"Statistical Approach for Text and Non-text Classifier in Off-line Handwritten Document","authors":"B. Pravalpruk, S. Watcharabutsarakham","doi":"10.1109/ecti-con49241.2020.9158334","DOIUrl":"https://doi.org/10.1109/ecti-con49241.2020.9158334","url":null,"abstract":"Hand writing and hand drawing are natural ways to take note. A pen and papers are used to make a note for a long time. In digital era, the notes are often converted into a durable and formal format for further use. Therefore, the conversion application was developed in many fields with many skill such as handwritten recognition, object recognition, object classification, and others. In this paper, we demonstrate a method to classify connected components as flowchart and text. We use the Online Handwritten Flowchart Dataset (OHFD) which contained 419 handwritten flowcharts to benchmark our methodology. The result shown our classification technique get F1-score 77.6%.","PeriodicalId":371552,"journal":{"name":"2020 17th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology (ECTI-CON)","volume":"46 3 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":"116697703","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-con49241.2020.9158226
Prateep Manasummakij, Chatrpol Pakasiri
This paper presents a class-E power amplifier circuit designed using Thevenin impedance. The aim of the method is to maximize the output power of the circuit. The measured results showed that the proposed method yielded output power of 24.33 dBm comparing to 24.99 dBm of the simulated one.
{"title":"Maximized Output Power Design of Switch-Mode Power Amplifier with Thevenin Impedance","authors":"Prateep Manasummakij, Chatrpol Pakasiri","doi":"10.1109/ecti-con49241.2020.9158226","DOIUrl":"https://doi.org/10.1109/ecti-con49241.2020.9158226","url":null,"abstract":"This paper presents a class-E power amplifier circuit designed using Thevenin impedance. The aim of the method is to maximize the output power of the circuit. The measured results showed that the proposed method yielded output power of 24.33 dBm comparing to 24.99 dBm of the simulated one.","PeriodicalId":371552,"journal":{"name":"2020 17th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology (ECTI-CON)","volume":"27 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":"125854849","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-CON49241.2020.9158209
Widya Saputra, A. Muklason, Baiq Z.H. Rozaliya
In the literature, almost all optimization problems in NP-hard class are solved by meta-heuristics approach. However, this approach has the drawback of requiring tuning parameters for each different problem domain and different instances of the same problem. This approach is considered less effective in resolving these problems. Therefore, a new approach is needed, namely the hyper-heuristics approach that is able to solve cross-domain problems. Hyper-heuristic is one of the approximate search methods which is able to provide solutions to NP-hard problems in polynomial time, as well as giving fairly good and acceptable results. This method has two properties of search space, namely the selection of LLH and the acceptance of solutions (move acceptance). This approach works in barrier domains rather than directly working in problem domains. With these properties, hyper-heuristic is able to solve problems in different domains. In addition, hyper-heuristics has a learning mechanism through feedback from previously generated solutions. This final project tries to apply a hyperheuristic algorithm in six combinatorial optimization problem domains, namely SAT, Bin Packing, Flow Shop, Personnel Scheduling, TSP, and VRP. The method that will be used in this final project is Self Adaptive - Great Deluge (SAD-GED). The Self Adaptive mechanism is used to make LLH selection to be used, while the Great Deluge is used in determining the acceptance of solutions (move acceptance) in a hyperheuristic framework. The application of the SAD-GED algorithm is expected to be able to provide better results than the existing algorithm used previously, namely Simple Random - Simulated Annealing.
在文献中，几乎所有NP-hard类的优化问题都是用元启发式方法解决的。然而，这种方法的缺点是需要为每个不同的问题域和同一问题的不同实例调优参数。人们认为这种方法在解决这些问题方面效果较差。因此，需要一种新的方法，即能够解决跨领域问题的超启发式方法。超启发式是一种近似搜索方法，它能够在多项式时间内解决np困难问题，并给出相当好的和可接受的结果。该方法具有搜索空间的两个性质，即LLH的选择和解的接受(移动接受)。这种方法适用于障碍领域，而不是直接适用于问题领域。有了这些特性，超启发式能够解决不同领域的问题。此外，超启发式还具有通过先前生成的解决方案的反馈来学习的机制。这个最终的项目尝试在六个组合优化问题领域中应用超启发式算法，即SAT, Bin Packing, Flow Shop, Personnel Scheduling, TSP和VRP。在这个期末项目中使用的方法是自适应-大洪水(SAD-GED)。在超启发式框架中，自适应机制用于做出LLH选择，而大洪水机制用于确定解决方案的接受度(移动接受度)。应用SAD-GED算法有望提供比之前使用的简单随机模拟退火算法更好的结果。
{"title":"Self Adaptive Learning – Great Deluge Based Hyper-heuristics for Solving Cross Optimization Problem Domains","authors":"Widya Saputra, A. Muklason, Baiq Z.H. Rozaliya","doi":"10.1109/ECTI-CON49241.2020.9158209","DOIUrl":"https://doi.org/10.1109/ECTI-CON49241.2020.9158209","url":null,"abstract":"In the literature, almost all optimization problems in NP-hard class are solved by meta-heuristics approach. However, this approach has the drawback of requiring tuning parameters for each different problem domain and different instances of the same problem. This approach is considered less effective in resolving these problems. Therefore, a new approach is needed, namely the hyper-heuristics approach that is able to solve cross-domain problems. Hyper-heuristic is one of the approximate search methods which is able to provide solutions to NP-hard problems in polynomial time, as well as giving fairly good and acceptable results. This method has two properties of search space, namely the selection of LLH and the acceptance of solutions (move acceptance). This approach works in barrier domains rather than directly working in problem domains. With these properties, hyper-heuristic is able to solve problems in different domains. In addition, hyper-heuristics has a learning mechanism through feedback from previously generated solutions. This final project tries to apply a hyperheuristic algorithm in six combinatorial optimization problem domains, namely SAT, Bin Packing, Flow Shop, Personnel Scheduling, TSP, and VRP. The method that will be used in this final project is Self Adaptive - Great Deluge (SAD-GED). The Self Adaptive mechanism is used to make LLH selection to be used, while the Great Deluge is used in determining the acceptance of solutions (move acceptance) in a hyperheuristic framework. The application of the SAD-GED algorithm is expected to be able to provide better results than the existing algorithm used previously, namely Simple Random - Simulated Annealing.","PeriodicalId":371552,"journal":{"name":"2020 17th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology (ECTI-CON)","volume":"212 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":"123558085","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-CON49241.2020.9158227
A. Suchaimanacharoen, T. Kasetkasem, S. Marukatat, I. Kumazawa, P. Chavalit
With the assumption that capital markets follow the semi-strong form of efficient market hypothesis (EMH), numerous efforts have been taken to defeat the non-stationary financial market, ranging from time series analysis, artificial intelligence for prices prediction, to automated decision making by reinforcement learning. This experiment integrated the power of time series forecasting of neural network with the competence of actions selecting of the reinforcement learning. CNN was trained first to predict future prices, and then it fed the output to the policy gradient (PG) model together with historical data to empower the trading decisions. The experiment was conducted on 30 minutes interval of EUR/USD pair in Forex between 2014 and 2018. Our experimental results showed that our model can achieve higher return in both train and validate samples than buy and hold strategy.
{"title":"Empowered PG in Forex Trading","authors":"A. Suchaimanacharoen, T. Kasetkasem, S. Marukatat, I. Kumazawa, P. Chavalit","doi":"10.1109/ECTI-CON49241.2020.9158227","DOIUrl":"https://doi.org/10.1109/ECTI-CON49241.2020.9158227","url":null,"abstract":"With the assumption that capital markets follow the semi-strong form of efficient market hypothesis (EMH), numerous efforts have been taken to defeat the non-stationary financial market, ranging from time series analysis, artificial intelligence for prices prediction, to automated decision making by reinforcement learning. This experiment integrated the power of time series forecasting of neural network with the competence of actions selecting of the reinforcement learning. CNN was trained first to predict future prices, and then it fed the output to the policy gradient (PG) model together with historical data to empower the trading decisions. The experiment was conducted on 30 minutes interval of EUR/USD pair in Forex between 2014 and 2018. Our experimental results showed that our model can achieve higher return in both train and validate samples than buy and hold strategy.","PeriodicalId":371552,"journal":{"name":"2020 17th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology (ECTI-CON)","volume":"90 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":"116203653","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}
Knowledge graphs (KGs) have been utilized by various business fields. One example is Google that stores data in knowledge graphs for searching and retrieval tasks. Even though these graphs have reached an impressive size, they are far from completeness. Missing relations in knowledge graphs is a severe problem for algorithms that operate over knowledge graphs. There are many researchers trying to develop knowledge graph embedding methods so that they can handle different types of relations. ConvKB is one of the knowledge graph embedding methods that utilize convolution neural networks (CNN). However, this method lacks the ability to handle symmetric relations. Being inspired by this limitation, we would like to enhance this method by proposing ConvKB+, which is obtained by modifying ConvKB’s CNN structure and introducing an additional relation vector. Our experiment results show that our method outperforms ConvKB by achieving higher MRR on some symmetric relations of the WN18RR dataset.
{"title":"Enhancing CNN Based Knowledge Graph Embedding Algorithms Using Auxiliary Vectors: A Case Study of Wordnet Knowledge Graph","authors":"Chanathip Pornprasit, Pattararat Kiattipadungkul, Peeranut Duangkaew, Suppawong Tuarob, Thanapon Noraset","doi":"10.1109/ecti-con49241.2020.9158288","DOIUrl":"https://doi.org/10.1109/ecti-con49241.2020.9158288","url":null,"abstract":"Knowledge graphs (KGs) have been utilized by various business fields. One example is Google that stores data in knowledge graphs for searching and retrieval tasks. Even though these graphs have reached an impressive size, they are far from completeness. Missing relations in knowledge graphs is a severe problem for algorithms that operate over knowledge graphs. There are many researchers trying to develop knowledge graph embedding methods so that they can handle different types of relations. ConvKB is one of the knowledge graph embedding methods that utilize convolution neural networks (CNN). However, this method lacks the ability to handle symmetric relations. Being inspired by this limitation, we would like to enhance this method by proposing ConvKB+, which is obtained by modifying ConvKB’s CNN structure and introducing an additional relation vector. Our experiment results show that our method outperforms ConvKB by achieving higher MRR on some symmetric relations of the WN18RR dataset.","PeriodicalId":371552,"journal":{"name":"2020 17th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology (ECTI-CON)","volume":"48 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":"114417246","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-con49241.2020.9158324
{"title":"ECTI-CON 2020 Cover Page","authors":"","doi":"10.1109/ecti-con49241.2020.9158324","DOIUrl":"https://doi.org/10.1109/ecti-con49241.2020.9158324","url":null,"abstract":"","PeriodicalId":371552,"journal":{"name":"2020 17th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology (ECTI-CON)","volume":"134 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":"128541057","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-con49241.2020.9158099
Sudarat Saengkeaw
The traditional concept-effect relationship model (CER model) aims at finding the student’s suggestion to improve personalized learning outcomes. To provide more benefits to the instructor, we apply association rule mining to searching for interesting relationships among all students’ in- class testing scores. This approach enhances instructors to better understand student learning performance and improve the instructor’s course design. The experimental results on a computer data mining course have demonstrated feasibility of the approach and the mining results provide feedback for supporting instructors in the form of strong association rules, which is found to be very useful in practical applications.
{"title":"Application of Association Rule Mining with Concept-Effect Relationship Model for Learning Diagnosis","authors":"Sudarat Saengkeaw","doi":"10.1109/ecti-con49241.2020.9158099","DOIUrl":"https://doi.org/10.1109/ecti-con49241.2020.9158099","url":null,"abstract":"The traditional concept-effect relationship model (CER model) aims at finding the student’s suggestion to improve personalized learning outcomes. To provide more benefits to the instructor, we apply association rule mining to searching for interesting relationships among all students’ in- class testing scores. This approach enhances instructors to better understand student learning performance and improve the instructor’s course design. The experimental results on a computer data mining course have demonstrated feasibility of the approach and the mining results provide feedback for supporting instructors in the form of strong association rules, which is found to be very useful in practical applications.","PeriodicalId":371552,"journal":{"name":"2020 17th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology (ECTI-CON)","volume":"91 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":"124757214","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-con49241.2020.9158082
T. Jariyanorawiss, Wachira Chongburee
This paper presents a simulation result of human head exposure to 2.6 GHz, which is one of the bands used in the recently launched 5G mobile networks. The method adopted in the simulation is Finite-Difference Time-Domain (FDTD), which divides the computational domains into a physical and an artificial absorbing domains. The physical domain consists of a dipole antenna representing the mobile phone and a human head model created by a set of 53 layers from Magnetic Resonance Imaging (MRI). The artificial absorbing domain is a 3-D reflectionless boundary which can be implemented by using Perfectly Matched Layers (PML). Also, the Specific Absorption Rate (SAR) value is averaged over 1 gram of the head tissues when the dipole is placed in the range of 1-10 cm from the human head. Additionally, the total power absorption computed from the electric field is also reported. The results suggest that as the distance between the mobile phone set and the human head increases, SAR decreases monotonically and exponentially. Meanwhile, with operating frequency 2.6 GHz, the power absorption tends to decrease but possibly increases at some particular distance. The key result is that for a radiated power of 0.6 W, none of the distances under test deliver SAR value that meet the 1.6 W/kg of the FCC standard. The simulation results conclude that the radiated power of approximately 0.25 W assures the compliance with the FCC standard at the distance of 1 cm.
{"title":"A Report on Human Head Exposure to a 2.6 GHz Mid-Band of 5G by Using FDTD Method","authors":"T. Jariyanorawiss, Wachira Chongburee","doi":"10.1109/ecti-con49241.2020.9158082","DOIUrl":"https://doi.org/10.1109/ecti-con49241.2020.9158082","url":null,"abstract":"This paper presents a simulation result of human head exposure to 2.6 GHz, which is one of the bands used in the recently launched 5G mobile networks. The method adopted in the simulation is Finite-Difference Time-Domain (FDTD), which divides the computational domains into a physical and an artificial absorbing domains. The physical domain consists of a dipole antenna representing the mobile phone and a human head model created by a set of 53 layers from Magnetic Resonance Imaging (MRI). The artificial absorbing domain is a 3-D reflectionless boundary which can be implemented by using Perfectly Matched Layers (PML). Also, the Specific Absorption Rate (SAR) value is averaged over 1 gram of the head tissues when the dipole is placed in the range of 1-10 cm from the human head. Additionally, the total power absorption computed from the electric field is also reported. The results suggest that as the distance between the mobile phone set and the human head increases, SAR decreases monotonically and exponentially. Meanwhile, with operating frequency 2.6 GHz, the power absorption tends to decrease but possibly increases at some particular distance. The key result is that for a radiated power of 0.6 W, none of the distances under test deliver SAR value that meet the 1.6 W/kg of the FCC standard. The simulation results conclude that the radiated power of approximately 0.25 W assures the compliance with the FCC standard at the distance of 1 cm.","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":"130659643","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-con49241.2020.9158133
Krin Chinprasatsak, N. Niparnan, A. Sudsang
This research compares 4 neural networks from the original researches (I. Backpropagation Neural Network II. Bayesian Regularized Neural Network III. Empirical Mode Decomposition Stochastic Time Strength Neural Network IV. Random Data-time Effective Radial Basis Function Neural Network) and 2 proposed neural networks (I. Empirical Mode Decomposition Random Data-time Effective Radial Basis Function Neural Network II. Empirical Mode Decomposition Random Data-time Effective Bayesian Regularized Neural Network) for predicting the exchange rate of EUR/USD currency pairs using input as a technical indicator and evaluating the networks with trading simulations consisting of investment strategies, risk management methods and financial management principles. The experiments show that the proposed neural networks yield higher returns than the original researches.
{"title":"Neural Network for Forecasting High Price and Low Price on Foreign Exchange Market","authors":"Krin Chinprasatsak, N. Niparnan, A. Sudsang","doi":"10.1109/ecti-con49241.2020.9158133","DOIUrl":"https://doi.org/10.1109/ecti-con49241.2020.9158133","url":null,"abstract":"This research compares 4 neural networks from the original researches (I. Backpropagation Neural Network II. Bayesian Regularized Neural Network III. Empirical Mode Decomposition Stochastic Time Strength Neural Network IV. Random Data-time Effective Radial Basis Function Neural Network) and 2 proposed neural networks (I. Empirical Mode Decomposition Random Data-time Effective Radial Basis Function Neural Network II. Empirical Mode Decomposition Random Data-time Effective Bayesian Regularized Neural Network) for predicting the exchange rate of EUR/USD currency pairs using input as a technical indicator and evaluating the networks with trading simulations consisting of investment strategies, risk management methods and financial management principles. The experiments show that the proposed neural networks yield higher returns than the original researches.","PeriodicalId":371552,"journal":{"name":"2020 17th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology (ECTI-CON)","volume":"171 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":"124169958","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：