Transferring topical knowledge from auxiliary long texts for short text clustering

Abstract

With the rapid growth of social Web applications such as Twitter and online advertisements, the task of understanding short texts is becoming more and more important. Most traditional text mining techniques are designed to handle long text documents. For short text messages, many of the existing techniques are not effective due to the sparseness of text representations. To understand short messages, we observe that it is often possible to find topically related long texts, which can be utilized as the auxiliary data when mining the target short texts data. In this article, we present a novel approach to cluster short text messages via transfer learning from auxiliary long text data. We show that while some previous works for enhancing short text clustering with related long texts exist, most of them ignore the semantic and topical inconsistencies between the target and auxiliary data and may hurt the clustering performance on the short texts. To accommodate the possible inconsistencies between source and target data, we propose a novel topic model -Dual Latent Dirichlet Allocation (DLDA) model, which jointly learns two sets of topics on short and long texts and couples the topic parameters to cope with the potential inconsistencies between data sets. We demonstrate through large-scale clustering experiments on both advertisements and Twitter data that we can obtain superior performance over several state-of-art techniques for clustering short text documents.

Investigating Online Destination Images Using a Topic-Based Sentiment Analysis Approach

“…First, the LDA model is based on the assumption that each tourist review is assumed to be a "bag-of-words" document. LDA may not obtain a desirable performance when applied to short documents [60]. Second, selecting a globally optimal topic number to train the LDA model is difficult.…”

FSFP: Transfer Learning From Long Texts to the Short

“…Although rarely, there also some people did the work that transfer methods or skills form the long text to the short. For example, Jin et al [11] proposed a DLDA model. It extracted two sets of topics from the source and target domains and used a binary switch variable to control the forming process of the documents.…”

Vector Representation of Words for Detecting Topic Trends over Short Texts

“…Some researchers aim to aggregate a subset of short texts to form a long texts. And then topic models are applied over these long texts [7]. Other ingenious methods, such as MA-LDA [8] and MB-LDA [9], take some features into consideration and incorporate these features into LDA.…”