Abstract:Social media produces large amounts of contents every day. How to predict the potential influences of the contents from a social reply feedback perspective is a key issue that has not been explored. Thus, we propose a novel task named reply keyword prediction in social media, which aims to predict the keywords in the potential replies as many aspects as possible. One prerequisite challenge is that the accessible social media datasets labeling such keywords remain absent. To solve this issue, we propose a new dataset, to study the reply keyword prediction in Social Media. This task could be seen as a single-turn dialogue keyword prediction for open-domain dialogue system. However, existing methods for dialogue keyword prediction cannot be adopted directly, which have two main drawbacks. First, they do not provide an explicit mechanism to model topic complementarity between keywords which is crucial in social media to controllably model all aspects of replies. Second, the collocations of keywords are not explicitly modeled, which also makes it less controllable to optimize for fine-grained prediction since the context information is much less than that in dialogue. To address these issues, we propose a two-stage disentangled framework, which can optimize the complementarity and collocation explicitly in a disentangled fashion. In the first stage, we use a sequence-to-set paradigm via multi-label prediction and determinantal point processes, to generate a set of keyword seeds satisfying the complementarity. In the second stage, we adopt a set-to-sequence paradigm via seq2seq model with the keyword seeds guidance from the set, to generate the more-fine-grained keywords with collocation. Experiments show that this method can generate not only a more diverse set of keywords but also more relevant and consistent keywords. Furthermore, the keywords obtained based on this method can achieve better reply generation results in the retrieval-based system than others.

Seq2Set2Seq: A Two-stage Disentangled Method for Reply Keyword Generation in Social Media via Multi-label Prediction and Determinantal Point Processes

Generating Diverse Conversation Responses by Creating and Ranking Multiple Candidates

Who is Answering Whom? Finding "Reply-To" Relations in Group Chats with Deep Bidirectional LSTM Networks

Dual Semantic Knowledge Composed Multimodal Dialog Systems

Multi-task Prompt Words Learning for Social Media Content Generation

Response Enhanced Semi-supervised Dialogue Query Generation

SocialSift: Target Query Discovery on Online Social Media With Deep Reinforcement Learning

Sequence to Backward and Forward Sequences: A Content-Introducing Approach to Generative Short-Text Conversation.

Promoting Diversity for End-to-End Conversation Response Generation.

Predict-Then-Decide: A Predictive Approach for Wait or Answer Task in Dialogue Systems

EM Pre-training for Multi-party Dialogue Response Generation

Dynamic Knowledge Routing Network For Target-Guided Open-Domain Conversation

Joint Learning for Addressee Selection and Response Generation in Multi-Party Conversation

Towards Robust Online Dialogue Response Generation

Prediction, selection, and generation: a knowledge-driven conversation system

Who Is Answering to Whom? Finding “Reply-To” Relations in Group Chats with Long Short-Term Memory Networks

Jointly Learning Sentiment, Keyword and Opinion Leader in Social Reviews

Generative multi-round chat dialogue method and system and computer-readable storage medium

Response Ranking with Multi-types of Deep Interactive Representations in Retrieval-based Dialogues

Incorporating Social Role Theory into Topic Models for Social Media Content Analysis.

Keyword-Aware Transformers Network for Chinese Open-Domain Conversation Generation