Many online community question answering sites have a user registration function. By using the information entered by the user, it becomes possible for the user to provide the information contents he / she desires, and it is possible to lead the user to use the service for a long time by functions for improving their satisfaction. In addition, it is possible to maximize revenue of services by delivering personaliz ed advertisements. Information on important events (life events) occurring in a person's lifetime, particularly marriage or childbirth, is known as the time when the person collects various information to make decisions and tend to carry out purchase activ ities of high unit price. However, there are certain barriers for the user to input such life event information, and there are cases in which some users do not input their own perso nal information There are users who use the service without inputting esti mated date of confinement even at Mamari which is a community question answering site for women who solves the question of pregnancy, childbirth, and child rearing Therefore, w e constructed a system that predicts user's baby’s estimated date of confinemen t by machine learning from the question posted by the user without inputting estimated date of confinement.

View full abstract

Concerning the procurement of IT systems in the ministries and agencies of the central government, various problems have been indicated for a long time such as negotiated contract, one party bid, or vender lock in. Responding to these issues, the government established “the basic policy on the government IT procurement” and defined a standard procurement procedure to realize free, fair, and competitive procurement environment and to ensure transparency and fairness in the IT procurement process in 2007. Since then, the IT procurement by the ministries and agencies has been implemented according to this policy. Under such circumstances, all IT systems development were planned at once to start operating the systems in the ministries and agencies before implementing the social security and tax number (My Number) system from January, 2016. This study aims to clarify if the free, fair and competitive procurement environment has been realized in a way that the basic policy aimed originally, based on the actual IT procurement results by the ministries and agencies to execute the My Number system.

View full abstract

In the decision making of sales activities, the means for greatly increasing the efficiency of sales activities are required by eliminating individuals' factors such as experience and intuition In order to solve this problem we are developing the business decision support system using a machine learning model. For that, it is necessary to learn the process of sales activities with high order acceptance probability, so the process discovery technology that extracts regularity from the decision making is essential. In the process discovery method in conventional process mining the regularity and rules of business process are extracted from information on execution order of events contained in the event logs output from the systems How ever in the decision making of sales activities it is difficult to apply the conventional process discovery method because the rules are not known in advance and the input information is unstructured data such as business diaries. In this paper we provid e the process discovery method using the activity estimation based on unstructured data and the process estimation for stochastic expression of regularity in an atypical process for the process discovery method of decision making in sales activities.

View full abstract