Indonesia Conference Directory

Corresponding Author
Maju Sumanto

Institutions
Bandung Institute of Technology (ITB) - Computational Science

Abstract
Machine Learning is an artificial intelligence system, where the system has the ability to learn automatically from experience without being explicitly programmed. The learning process from Machine Learning starts from observing the data and then looking at the pattern of the data. The main purpose of this process is to make computers learn automatically. In this study, we used machine learning to predict molecular atomization energy. We use two methods namely Neural Network and Extreme Gradient Boosting. Both methods have several parameters that must be adjusted so the predicted value of the atomization energy of the molecule has the lowest possible error. We are trying to find the right parameter values for both methods. For the neural network method, it is quite difficult to find the right parameter value because it takes a long time to train the model of the neural network to find out whether the model is good or bad, while for the Extreme Gradient Boosting method the time needed to train the model is shorter, so it is quite easy to find the right parameter values for the model. This study also looked at the effects of the modification on the dataset with the output transformation of normalization and standardization then removing molecules containing Br atoms and changing the entry in the Coulomb matrix to 0 if the distance between atoms in the molecule exceeds 2 angstrom.

Keywords
Machine Learning, Neural Network, Extreme Gradient Boosting, Atomization Energy, Molecule.

Topic
Artificial Intelligent and soft computing

Link: https://ifory.id/abstract/ymfUR6MdYA4x

Corresponding Author
Danial -

Institutions
Direktorat Sistem dan Teknologi Informasi, Institut Teknologi Bandung
Jalan Ganesa 10, Bandung 40132, Indonesia

Abstract
In Indonesia, bandwidth usage management in higher education is still a great challenge. Every year, the internet bandwidth and the budget are higher, but the growth of scientific publication to the cumulative universities is still under the other countries in ASEAN. In the other side, the universities have not measure the effective of usage bandwidth for their activities. The log data is managed rarely. The user and data can be clustered. So the management have the picture for their bandwidth policies. In this paper, there will be comparing the clustering that using Agglomerate, K-Means, and K-Means algorithms. The results is to find the best clustering and make a pattern for ITB network usage. The method is implemented in real data in ITB network. The campus management will have a good data and information for their policies of bandwidth management.

Keywords
Clustering, Bandwidth Management, Data Management, Internet

Topic
Artificial Intelligent and soft computing

Link: https://ifory.id/abstract/XvGqJ4FCWVux

Corresponding Author
Joshua Dwi Prasetyo

Institutions
Internet of Things Laboratory, Department of Physics, Institut Teknologi Bandung

Abstract
Increasing demand and use of energy and declining energy source make today’s society have to do energy efficiency. The main focus of this energy efficiency research is AC (Air Conditioner) system because it contributes the most of total power consumption of building. The problems hindering the energy efficiency process on AC system is that the sensor in the AC system doesn’t represent the room’s thermal condition, parameters relevant to the thermal condition are not considered, and the control used only uses temperature optimization without considering energy efficiency. The solution proposed is designing a system that can overcome the problems with Internet of Things (IoT) technology. Other than physics modelling on thermal characteristics using thermodynamics and heat transfer concepts, Artificial Neural Network is also used to manage control part of the system. This research is conducted using experiment method on AC system and room of IoT Laboratory, Department of Physics, Institut Teknologi Bandung. The study on the effects of Duty Cycle (DC) and difference between outdoor and indoor temperature on thermal characteristics value, time needed to obtain thermal characteristics, the effects of DC and outdoor temperature on indoor temperature, the effects of room occupancy on thermal characteristics value, and control result have been conducted. The control has been tested: the average temperature is 21.98$^0$C on outdoor temperature value about 27$^0$C, the temperature range is 21.98$^0$C $pm$ 1.7$^0$C, and the saving obtained is 60%.

Keywords
Energy efficiency, AC system, thermal condition, IoT, Artificial Neural Network

Topic
Artificial Intelligent and soft computing

Link: https://ifory.id/abstract/qP2Ha3ptUWnz

Corresponding Author
Joshua Dwi Prasetyo

Institutions
Internet of Things Laboratory, Department of Physics, Institut Teknologi Bandung

Abstract
Energy efficiency is an effort to reduce energy use in doing work. One technology that can be used for energy efficiency is the Internet of Things (IoT). In this study, the energy savings of the room cooling system were reviewed. In order to achieve energy savings in the room cooling system, it takes an room cooling control that can perform energy efficiency while still considering thermal comfort. In this research, a study of the thermal characteristics of the room using IoT technology was carried out. From the results of this study, thermal parameters can be obtained from the room. This parameters is then processed into input variable to obtain the room cooling system control function using Machine Learning. A cyber physical system modeling has been carried out on room cooling systems with IoT technology. This system is able to get the value of room temperature, room humidity, ambient temperature, and the use of room cooling power and store them in Big Data. Energy efficiency control is done by using Pulse Width Modulation (PWM) method on compressor unit. In this paper, the performance of the control system is studied by variating the control period and set point temperature. Energy efficiency by the control reaches 73\% on outdoor temperature about 30$^0$C, the indoor temperature range obtained is $pm$ 0.4$^0$C from set point temperature value, and the temperature results achieved by the control has 0.02$^0$C difference with the temperature desired by the user.

Keywords
Energy efficiency, room cooling system, IoT, thermal characteristics, Machine Learning

Topic
Artificial Intelligent and soft computing

Link: https://ifory.id/abstract/BhUQLZDNJ7Xv