Parameterized Convolutional Neural Networks for Aspect Level Sentiment Classification

Huang, Binxuan; Carley, Kathleen M.

doi:10.18653/v1/d18-1136

Cited by 155 publications

(76 citation statements)

References 17 publications

(18 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Geotagged content on social media is also a direct way to understand events and the corresponding regional users' opinions [18], [19]. For example, Sakaki et al used geotagged tweets to detect earthquakes in real-time [20].…”

Section: A the Use Of Geotagsmentioning

confidence: 99%

A large-scale empirical study of geotagging behavior on Twitter

Huang

Carley

2019

Proceedings of the 2019 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining

Self Cite

View full text Add to dashboard Cite

Geotagging on social media has become an important proxy for understanding people's mobility and social events. Research that uses geotags to infer public opinions relies on several key assumptions about the behavior of geotagged and non-geotagged users. However, these assumptions have not been fully validated. Lack of understanding the geotagging behavior prohibits people further utilizing it. In this paper, we present an empirical study of geotagging behavior on Twitter based on more than 40 billion tweets collected from 20 million users. There are three main findings that may challenge these common assumptions. Firstly, different groups of users have different geotagging preferences. For example, less than 3% of users speaking in Korean are geotagged, while more than 40% of users speaking in Indonesian use geotags. Secondly, users who report their locations in profiles are more likely to use geotags, which may affects the generability of those location prediction systems on non-geotagged users. Thirdly, strong homophily effect exists in users' geotagging behavior, that users tend to connect to friends with similar geotagging preferences.

show abstract

Section: A the Use Of Geotagsmentioning

confidence: 99%

A large-scale empirical study of geotagging behavior on Twitter

Huang

Carley

2019

Proceedings of the 2019 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining

Self Cite

View full text Add to dashboard Cite

show abstract

“…Independently, there exists several works in literature focused on developing supervised and unsupervised models for understanding sentiment from user utterance [32][33][34]. However, there exists very little work that utilizes these additional information of the user behavior in the decision making process for the VA to be efficient and competent enough to converse and execute its goal appropriately.…”

Section: Plos Onementioning

confidence: 99%

Towards sentiment aided dialogue policy learning for multi-intent conversations using hierarchical reinforcement learning

Saha

Bhattacharyya

2020

PLoS ONE

View full text Add to dashboard Cite

Developing a Dialogue/Virtual Agent (VA) that can handle complex tasks (need) of the user pertaining to multiple intents of a domain is challenging as it requires the agent to simultaneously deal with multiple subtasks. However, majority of these end-to-end dialogue systems incorporate only user semantics as inputs in the learning process and ignore other useful user behavior and information. Sentiment of the user at the time of conversation plays an important role in securing maximum user gratification. So, incorporating sentiment of the user during the policy learning becomes even more crucial, more so when serving composite tasks of the user. Methodology As a first step towards enabling the development of sentiment aided VA for multi-intent conversations, this paper proposes a new dataset, annotated with its corresponding intents, slot and sentiment (considering the entire dialogue history) labels, named SentiVA, collected from open-sourced dialogue datasets. In order to integrate these multiple aspects, a Hierarchical Reinforcement Learning (HRL) specifically options based VA is proposed to learn strategies for managing multi-intent conversations. Along with task success based immediate rewards, sentiment based immediate rewards are also incorporated in the hierarchical value functions to make the VA user adaptive. Findings Empirically, the paper shows that task based and sentiment based immediate rewards cumulatively are required to ensure successful task completion and attain maximum user satisfaction in a multi-intent scenario instead of any of these rewards alone. Practical implications The eventual evaluators and consumers of dialogue systems are users. Thus, to ensure a fulfilling conversational experience involving maximum user satisfaction requires VA to consider user sentiment at every time-step in its decision making policy.

show abstract

“…Similarly, to perform ATSA the model is extended with third CNN that extracts contextual information of aspect terms. Along the same line of using CNN, Huang et al [147] introduced two novel CNNs named CNN-parameterized filters (CNN-PF) and CNN-parametrized gate (CNN-PG). The CNN-PF uses CNN to extract information from the sentence and uses the aspect specific features as parameters to the CNN.…”

Section: Applicationmentioning

confidence: 99%

“…Second, the significance of attention mechanism is manifested in all datasets and tasks of ABSA. This is evidenced by the fact that all models on ABSA are attention-based, except GCAE [30] and PG-CNN/PF-CNN [147] models, which do not contain attention mechanism. Furthermore, TNet [123] has proved the power of combining Bi-LSTM that deals with long-term dependencies and CNN that learns local sentiment information.…”

Section: Performance Of DL Models On Aspect-based Sentiment Analysismentioning

confidence: 99%