Time Aware Knowledge Extraction for Microblog Summarization on Twitter

Microblogging services like Twitter and Facebook collect millions of user generated content every moment about trending news, occurring events, and so on. Nevertheless, it is really a nightmare to find information of interest through the huge amount of available posts that are often noise and redundant. In general, social media analytics services have caught increasing attention from both side research and industry. Specifically, the dynamic context of microblogging requires to manage not only meaning of information but also the evolution of knowledge over the timeline. This work defines Time Aware Knowledge Extraction (briefly TAKE) methodology that relies on temporal extension of Fuzzy Formal Concept Analysis. In particular, a microblog summarization algorithm has been defined filtering the concepts organized by TAKE in a time-dependent hierarchy. The algorithm addresses topic-based summarization on Twitter. Besides considering the timing of the concepts, another distinguish feature of the proposed microblog summarization framework is the possibility to have more or less detailed summary, according to the user’s needs, with good levels of quality and completeness as highlighted in the experimental results.

💡 Research Summary

The paper introduces a novel framework called Time Aware Knowledge Extraction (TAKE) for summarizing microblog streams, particularly Twitter, by jointly modeling temporal dynamics and semantic content. Traditional microblog summarization approaches either treat time and meaning in separate stages (e.g., LDA‑based topic models followed by temporal post‑processing) or focus solely on one aspect, which limits their ability to capture the evolution of a story as it unfolds. TAKE addresses this limitation by extending Fuzzy Formal Concept Analysis (FFCA) with a temporal dimension, thereby constructing a “time‑fuzzy lattice” that simultaneously encodes when tweets appear and what concepts they convey.

The framework consists of three main phases. First, Microblog Content Analysis extracts features from raw tweets. Each tweet is “wikified,” i.e., mapped to a set of Wikipedia entities that represent its semantic content. Simultaneously, the Offline Peak‑Finding Algorithm (OPAD) identifies temporal peaks in tweet volume, which serve as coarse time slots reflecting bursts of activity around an event. Second, Time‑Aware Knowledge Extraction (TAKE) takes the wikified tweets and their timestamps as input to build a fuzzy formal context K = (G, M, I). Here G is the set of tweets, M the set of Wikipedia entities, and I a fuzzy relation assigning a membership value µ(g,m)∈