Yoshiyuki Mizuno

SVM integrated case based restarting GA for further improving solar flare prediction

Solar activity has various influences on the global environment. Specifically, it may have serious impacts on the Earth such as satellite damage, etc. and power plant failures causing more serious disaster. For a precise forecast of larger scale solar flares causing serious disaster, it is important to improve the space weather forecast, a daily forecast of the solar flare. In our work so far, a machine-learning algorithm called Support Vector Machine (SVM) was used. We extended this technology by integrating Case Based Genetic Algorithm (CBGA) for a more precise forecast. It was shown experimentally that triple mutation rate on the slowdown of evolution in our CBGA improves considerably (e.g. another 5%) more than original mutation rate in the True Skill Statistics TSS. For further obtaining the optimality towards more imbalanced data analysis applicable to the recognition of serious disaster or medical disease, Restart CBGA is proposed with its expected effect. Here GA integrating SVM is restarted using highly optimized but diversified solutions in the case base as initial individuals. Further this restart CBGA is repetitively and evolutionary performed, evolving and maintaining the case base by the result of each (restarted) GA.

Solar flare prediction by SVM integrated GA

Solar flare has various influences on the global environment, in particular on the magnetic storm and the likelihood of natural disasters. Specifically, it may have serious impacts on the Earth such as failure of satellite communication and navigation (GPS), satellite damage, increased radiation exposure to astronauts, geomagnetic storm and aurora, and power plant failures causing more serious disaster. For a precise forecast of larger scale solar flares causing serious disaster, it is important to improve the space weather forecast, which is basically a daily forecast of the solar flare. In the work so far, a machine-learning algorithm called Support Vector Machine (SVM) was used to forecast the space weather. We extend this technology by integrating Genetic Algorithm (GA) elaborately combined with Case Based Reasoning for more precise forecast or imbalanced data classification. Finally, basic evaluation of this architectural idea called CBGALO shows it is promising in improving solar flare prediction.

An Adaptive System for Optimal Matches between Human Needs and Offers

The paper presents a very general and many purpose technique to represent human needs and offers along with a technology to find optimal matches. Moreover, the system is able to learn from its use by collecting user feedback and changing its parameters accordingly. This way, the system adjusts itself to the human expectations and desires and even follows the trend of these desires and expectations.

Solar flare prediction by SVM integrated CBGA with dynamic mutation rate

Solar activity has various influences on the global environment, in particular on the magnetic storm and the likelihood of natural disasters. Specifically, it may have serious impacts on the Earth such as failure of satellite communication and navigation (GPS), satellite damage, increased radiation exposure to astronauts, geomagnetic storm and aurora, and power plant failures causing more serious disaster. For a precise forecast of larger scale solar flares causing serious disaster, it is important to improve the space weather forecast, which is basically a daily forecast of the solar flare. In our work so far, a machine-learning algorithm called Support Vector Machine (SVM) was used to forecast the space weather. Here, we propose to extend this technology by integrating Case Based Genetic Algorithm (CBGA) for a more precise forecast and present an evaluation of this approach. Experimental evaluation shows that triple mutation rate on the slowdown of evolution in our Genetic Algorithm improves considerably (e.g. another 5%) more than original mutation rate in the True Skill Statistics TSS.

Improvement of Sun Flare Prediction by SVM Integrated GA

Solar activity has various influences on the global environment, in particular on the weather and the likelihood of natural disasters. In particular, it may have serious impacts on Earth such as failure of satellite communication and navigation (GPS), satellite damage, increased radiation exposure to astronauts, geomagnetic storm and aurora, and power plant failures causing more serious disaster. For a precise forecast of larger scale solar flares causing serious disaster, it is important to improve the space weather forecast, which is basically a daily forecast of the solar flare. In our work so far, a machine-learning algorithm called Support Vector Machine (SVM) was used to forecast the space weather. Here, we propose to extend this technology by integrating a Genetic Algorithm (GA) for a more precise forecast and present an evaluation of this approach.

A Sale-oriented Online-shop Management Support Method for E-commerce

In online business, it is important to construct sale web pages offering attractive services for popular products in order to improve page access and purchase rates. Moreover, online shop owners are required to hold various types of sales frequently throughout the year to keep customers coming back. However, along with these web pages, management of product DB (Data Base) is required to set up complicated sales contents with high efficiency and reliability. To satisfy this requirement, a sale-oriented online-shop management support method is proposed. In this method, using a sale management web page, online-shop owners can efficiently and reliably construct attractive sale pages consistent with their DB, using hierarchical sale knowledge created by a software agent and modified by them. Namely, the automatic construction and interactive modifications of sale pages as well as the automatic/interactive update of DB for each sale product group can be done consistently and dynamically, synchronized with the sale.