Lead User Identification In Online Innovation Communities

Under the circumstance of open innovation, users, especially lead users are playing a more and more important role in modern innovation. They not only provide consumer needs information for enterprises, but also participate in the development and improvement of products.Lead user innovation can not only help enterprises improve product development efficiency and save costs, but also enable them to meet the customers' customized needs. Therefore, it is particularly necessary to incorporate lead users into enterprises' innovation system.Before the use of lead users to participate in innovation, the identification of lead users is the first task. Since Von Hippel E put forward the concept of lead user, the identification of lead users has become a hot topic for many scholars. Although there are already some methods of lead user identification in the current research, these methods are inefficient and cost a lot of manpower and time. The rapid development of Internet technology has enabled enterprises to establish their own online innovation communities, thus to provide a new platform for user innovation and provided new perspectives of lead user identification. Many scholars began to focus on the user data in online innovation communities.Based on the existing literature of user identification and related methods, a set of lead user identification method based on supervised machine learning classification algorithm is put forward focusing on user data in online innovation communities. First, the significant influence of the content of innovation, active, community influence and users' relationship on users' leading status is found through literature reviews, expert interviews, senior user visits and the long-term observation of some mainstream online innovation communities, and then an indicator system of lead user identification based on the user content information characteristics and user behavior data characteristics is built accordingly. Second, a model of lead user identification based on supervised machine learning classification algorithm is proposed. Finally, with the data of MIUI Forum, an experiment is taken as an example to test the model. The result of the experiment shows that there is a significant distinction of identification index proposed between lead users and non-lead users, and the model trained with real data is effective.