Generic process for extracting user profiles from social media using hierarchical knowledge bases

Abstract

We present the design and application of a generic approach for semantic extraction of professional interests from social media using a hierarchical knowledge-base and spreading activation theory. By this, we can assess to which extend a user's social media life reflects his or her professional life. Detecting named entities related to professional interests is conducted by a taxonomy of terms in a particular domain. It can be assumed that one can freely obtain such a taxonomy for many professional fields including computer science, social sciences, economics, agriculture, medicine, and so on. In our experiments, we consider the domain of computer science and extract professional interests from a user's Twitter stream. We compare different spreading activation functions and metrics to assess the performance of the obtained results against evaluation data obtained from the professional pUblications of the Twitter users. Besides selected existing activation functions from the literature, we also introduce a new spreading activation function that normalizes the activation w.r.t. to the outdegree of the concepts.

Influence of time on user profiling and recommending researchers in social media

“…We assume that a user's social media profile reflects the content of publications. This experiment extends our prior work [8] by analyzing whether older publications boost the similarity scores and investigating the influence of the number of publications and number of tweets. The second experiment recommends related researchers profiled by their publications based on a user's social media profile.…”

“…First, we briefly present how to extract entities from texts (i .e., social media items and publications), following previous work [8]. Subsequently, we list the entity scoring functions applied in this paper in Section 3.2.…”

“…This paper extends our prior work [8] by conducting two experiments to assess different existing and new entity scoring functions and investigate the temporal influence of the used data. We employ a domain-specific knowledge base to reduce noise that Abel et al [3] observed.…”

“…They employed different scoring functions that reveal user interests which are not mentioned directly in Twitter and gave a score. Our previous work [8] provided approaches to extract professional interests from social media activities and publications, using a domain-specific knowledge base and spreading activation functions.…”

Inferring user interests in microblogging social networks: a survey

“…Predicting user interests Kapanipathi et al (2014), Kang and Lee (2016), Michelson and Macskassy (2010) Piao andBreslin (2016a,b,c,d, 2017a,b) Publication recommendations Nishioka and Scherp (2016), Große-Bölting et al (2015) Tweet recommendations Lu et al (2012), Sang et al (2015), Karatay and Karagoz (2015) Table 3 is a conceptual framework for discussing user modeling strategies proposed in the related work and to act as a "guide" to the rest of this survey. The rest of this paper is organized as follows.…”