Construction And Analysis Of The Network Of Scholars' Research Interest

There are some problems such as information overload and information islanding in the academic field.Most scholars do not have enough time to study the growing academic literature,only by focusing on core journals,core figures and conferences to understand the research trends in their academic research field.But this is likely to cause information island and they can't find the articles they are interested in and the scholars who share the same research interests with them.This paper is inspired by the user interest model,which is widely used by the electronic business platform and takes the geological field as an example to build the scholars' research interest network.We use topic model to extract scholars' research interest and combine different user interest model building methods to build the network.In order to ensure the quality of extracted interest subjects,we use the name disambiguation rules to identify authors,and through the domain dictionary construction,fixed phrase extraction to ensure the quality of Chinese word segmentation.On the one hand,we use LDA topic model to extract scholar research interest for collective scholar research interest network.On the other hand,we build individual scholar research interest network by calculating interest subject similarity.And then we use the methods of complex network and bipartite network to analyze the two networks.Geological research focus on oil and gas resources and mineral resources development.Finally,by extracting the dynamic evolution trend of scholars' research interest,this paper constructs a dynamic network of scholars' research interest.The construction of scholars' research interest network is the basis of academic recommendation system.On this basis,we can further explore the literature recommended research,scholars recommend research and so on.