No abstract

In times of a disaster, the information available on social media can be useful for several humanitarian tasks as disseminating messages on social media is quick and easily accessible. Disaster damage assessment is inherently multi-modal, yet most existing work on damage identification has focused solely on building generic classification models that rely exclusively on text or image analysis of online social media sessions (e.g., posts). Despite their empirical success, these efforts ignore the multi-modal information manifested in social media data. Conventionally, when information from various modalities is presented together, it often exhibits complementary insights about the application domain and facilitates better learning performance. In this work, we present Crisis-DIAS, a multi-modal sequential damage identification, and severity detection system. We aim to support disaster management and aid in planning by analyzing and exploiting the impact of linguistic cues on a unimodal visual system. Through extensive qualitative, quantitative and theoretical analysis on a real-world multi-modal social media dataset, we show that the Crisis-DIAS framework is superior to the state-of-the-art damage assessment models in terms of bias, responsiveness, computational efficiency, and assessment performance.

Developing human-like conversational agents is a prime area in HCI research and subsumes many tasks. Predicting listener backchannels is one such actively-researched task. While many studies have used different approaches for backchannel prediction, they all have depended on manual annotations for a large dataset. This is a bottleneck impacting the scalability of development. To this end, we propose using semi-supervised techniques to automate the process of identifying backchannels, thereby easing the annotation process. To analyze our identification module's feasibility, we compared the backchannel prediction models trained on (a) manually-annotated and (b) semi-supervised labels. Quantitative analysis revealed that the proposed semi-supervised approach could attain 95% of the former's performance. Our user-study findings revealed that almost 60% of the participants found the backchannel responses predicted by the proposed model more natural. Finally, we also analyzed the impact of personality on the type of backchannel signals and validated our findings in the user-study.

Consumer reviews online may contain suggestions useful for improving the target products and services. Mining suggestions is challenging because the field lacks large labelled and balanced datasets. Furthermore, most prior studies have only focused on mining suggestions in a single domain. In this work, we introduce a novel up-sampling technique to address the problem of class imbalance, and propose a multi-task deep learning approach for mining suggestions from multiple domains. Experimental results on a publicly available dataset show that our up-sampling technique coupled with the multi-task framework outperforms state-of-the-art open domain suggestion mining models in terms of the F-1 measure and AUC.

Temporal psycholinguistics can play a crucial role in studying expressions of suicidal intent on social media. Current methods are limited in their approach in leveraging contextual psychological cues from online user communities. This work embarks in a novel direction to explore historical activities of users and homophily networks formed between Twitter users for extracting suicidality trends. Empirical evidence proves the advantages of incorporating historical user profiling and temporal graph convolutional modeling for automated detection of suicidal connotations on Twitter. Related Work Challenges on Social MediaThe growth of social media websites hosts a number of challenges such as cyberbullying, suicide pacts, and radicalism that motivate suicidal behavior and P. Mathur and R. Sawhney-Equal contribution