Sentence Embedding Based Semantic Clustering Approach for Discussion Thread Summarization

Abstract

Huge data on the web come from discussion forums, which contain millions of threads. Discussion threads are a valuable source of knowledge for Internet users, as they have information about numerous topics. The discussion thread related to single topic comprises a huge number of reply posts, which makes it hard for the forum users to scan all the replies and determine the most relevant replies in the thread. At the same time, it is also hard for the forum users to manually summarize the bulk of reply posts in order to get the gist of discussion thread. Thus, automatically extracting the most relevant replies from discussion thread and combining them to form a summary are a challenging task. With this motivation behind, this study has proposed a sentence embedding based clustering approach for discussion thread summarization. The proposed approach works in the following fashion: At first, word2vec model is employed to represent reply sentences in the discussion thread through sentence embeddings/sentence vectors. Next, K-medoid clustering algorithm is applied to group semantically similar reply sentences in order to reduce the overlapping reply sentences. Finally, different quality text features are utilized to rank the reply sentences in different clusters, and then the high-ranked reply sentences are picked out from all clusters to form the thread summary. Two standard forum datasets are used to assess the effectiveness of the suggested approach. Empirical results confirm that the proposed sentence based clustering approach performed superior in comparison to other summarization methods in the context of mean precision, recall, and F-measure.

Exploring Community Awareness of Mangrove Ecosystem Preservation through Sentence-BERT and K-Means Clustering

“…Tokenization is the procedure for segmenting sentences into words, the results of which are called tokens [21]. Sentences are truncated based on whitespaces such as tabs and blanks, and punctuation symbols such as commas, semicolons, periods, and colons [22]. This stage eliminates unnecessary punctuation, spaces, and characters.…”

Facial expressions and the business cycle

“…In the appendix we provide a detailed description of how we implemented this additional filter, along with a sample of text ranked from least to most related to economic concepts (Table 12). A number of recent works suggest using a similar approach where a final computational step is taken to classify text data, see Bao et al (2020), Wu et al (2020), Wan et al (2020) and Khan et al (2020). A small scale evaluation of the classification strategy (100 pictures and associated text) results in an estimated precision of 0.78 and estimated recall of 0.80.…”

Summarizing twitter posts regarding COVID-19 based on n-grams

“…The domain of summarization is diverse in the scenario of different applications. One of the earliest approach focuses on summarizing a single document [18] then extended to summarize multiple documents [19], email thread [20], recognize the specific online arguments and dialogues [21], [22], and timeline summarization [23]. Comments summarization in social media concentrates on determining which posts are most relevant to a particular topic.…”