Interactive Search and Exploration in Discussion Forums Using Multimodal Embeddings

Abstract

In this paper we present a novel interactive multimodal learning system, which facilitates search and exploration in large networks of social multimedia users. It allows the analyst to identify and select users of interest, and to find similar users in an interactive learning setting. Our approach is based on novel multimodal representations of users, words and concepts, which we simultaneously learn by deploying a general-purpose neural embedding model. We show these representations to be useful not only for categorizing users, but also for automatically generating user and community profiles. Inspired by traditional summarization approaches, we create the profiles by selecting diverse and representative content from all available modalities, i.e. the text, image and user modality. The usefulness of the approach is evaluated using artificial actors, which simulate user behavior in a relevance feedback scenario. Multiple experiments were conducted in order to evaluate the quality of our multimodal representations, to compare different embedding strategies, and to determine the importance of different modalities. We demonstrate the capabilities of the proposed approach on two different multimedia collections originating from the violent online extremism forum Stormfront and the microblogging platform Twitter, which are particularly interesting due to the high semantic level of the discussions they feature. CCS CONCEPTS• Information systems → Multimedia and multimodal retrieval. KEYWORDS multimedia analytics, search, exploration, interactive learning, multimodal embeddings, online discussion forums, social multimedia • First, compact but meaningful multimodal content representations are needed to ensure the interactivity of the system

Towards Multi-Modal Conversational Information Seeking

“…From an algorithmic point-of-view, a major issue in MMIS relates to the fusion of several modalities (e.g., text and image) to obtain a meaningful representation. Recent state-of-the-art techniques employ joint representation techniques to find a latent space in which multiple modality information can be projected and compared [26]. This can be a challenging task, e.g., while content data such as text and images might be well-aligned, obtaining a latent space to perform joint content and user-preference representation (e.g., ratings, clicks, social-media information) might not be trivial.…”