Term-Community-Based Topic Detection with Variable Resolution
نویسندگان
چکیده
Network-based procedures for topic detection in huge text collections offer an intuitive alternative to probabilistic models. We present detail a method that is especially designed with the requirements of domain experts mind. Like similar methods, it employs community term co-occurrence graphs, but enhanced by including resolution parameter can be used changing targeted granularity. also establish ranking and use semantic word-embedding presenting communities way facilitates their interpretation. demonstrate application our widely corpus general news articles show results detailed social-sciences expert evaluations detected topics at various resolutions. A comparison Latent Dirichlet Allocation included. Finally, we discuss factors influence
منابع مشابه
Dark Web Portal Overlapping Community Detection Based on Topic Models
A hot research topic is the study and monitoring of online communities. Of course, homeland security institutions from many countries are using data mining techniques to perform this task, aiming to anticipate and avoid a possible menace to local peace. Tools such as social networks analysis and text mining have contributed to the understanding of these kinds of groups in order to develop count...
متن کاملCommunity Detection Based on Topic Distance in Social Tagging Networks
Research on the community detection in social tagging networks has attracted much attention in the last decade. Extracting the hidden topic information from tags provides a new way of thinking for community detection in social tagging networks. In this paper, a topic tagging network by extracting several topics from the tags through using the Latent Dirichlet Allocation (LDA) model is built fir...
متن کاملResolution limit in community detection.
Detecting community structure is fundamental for uncovering the links between structure and function in complex networks and for practical applications in many disciplines such as biology and sociology. A popular method now widely used relies on the optimization of a quantity called modularity, which is a quality index for a partition of a network into communities. We find that modularity optim...
متن کاملREINA at RepLab2013 Topic Detection Task: Community Detection
Social networks have become a large repository of comments which can extract multiple information. Twitter is one of the most widespread social networks and larger and is therefore an important source for detecting states of opinion, events and happenings before even the mainstream media. Topic detection is important to discover areas of interest that arise in the tweets. We have used classical...
متن کاملOverlapping Community Detection in Social Networks Based on Stochastic Simulation
Community detection is a task of fundamental importance in social network analysis. Community structures enable us to discover the hidden interactions among the network entities and summarize the network information that can be applied in many applied domains such as bioinformatics, finance, e-commerce and forensic science. There exist a variety of methods for community detection based on diffe...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Information
سال: 2021
ISSN: ['2078-2489']
DOI: https://doi.org/10.3390/info12060221