Edina: Building an Open Domain Socialbot with Self-dialogues

By Ben Krause, Marco Damonte, Mihai Dobre, Daniel Duma, Joachim Fainberg, Federico Fancellu, Emmanuel Kahembwe, Jianpeng Cheng, Bonnie Lynn Webber. 2017

44 citations recorded by the source corpus.

We present Edina, the University of Edinburgh's social bot for the Amazon Alexa Prize competition. Edina is a conversational agent whose responses utilize data harvested from Amazon Mechanical Turk (AMT) through an innovative new technique we call self-dialogues. These are conversations in which a single AMT Worker plays both participants in a dialogue. Such dialogues are surprisingly natural, efficient to collect and reflective of relevant and/or trending topics. These self-dialogues provide training data for a generative neural network as well as a basis for soft rules used by a matching score component. Each match of a soft rule against a user utterance is associated with a confidence score which we show is strongly indicative of reply quality, allowing this component to self-censor and be effectively integrated with other components. Edina's full architecture features a rule-based system backing off to a matching score, backing off to a generative neural network. Our hybrid data-driven methodology thus addresses both coverage limitations of a strictly rule-based approach and the lack of guarantees of a strictly machine-learning approach.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Natural Language Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

In Natural Language Processing

Citations in the corpus

Towards Exploiting Background Knowledge for Building Conversation Systems (2018, methodology): “In fact, our data collection method is inspired by the work of (Krause et al., 2017) where the authors use self-”
Conversational AI: The Science Behind the Alexa Prize (2018, background): “Socialbots received an obfuscated user hashcode to enable personalization for repeat users, and some teams, such as Edina (Damonte et al., 2017), added a level of personalization to their DM to track if a certain topic or type of response is doing particularly well with a user.”
Taskmaster-1: Toward a Realistic and Diverse Dialog Dataset (2019, background): “As Krause et al. (2017); Moghe et al.”
Towards Deep Conversational Recommendations (2018, methodology): “Krause et al. [7] introduce a dataset of self dialogues collected for the Amazon Alexa Prize competition 5 , using Amazon Mechanical Turk (AMT).”
BYU-EVE: Mixed Initiative Dialog via Structured Knowledge Graph Traversal and Conversational Scaffolding (2018, methodology): “This text may be retrieved from a curated and/or web-scraped repository [15, 24], generated via neural network models [15, 32], or constructed using a combination of templates and placeholders [15].”
Implicit Discourse Relation Identification for Open-domain Dialogues (2019, methodology): “We therefore constructed Edina-DR, the novel dataset of discourse relation pairs based on the publicly available self-dialogue Edina corpus which contains 24,165 multi-turn social conversations across 23 topics (Fainberg et al., 2018; Krause et al., 2017).”
TicketTalk: Toward human-level performance with end-to-end, transaction-based dialog systems (2020, methodology): “The TicketTalk movie ticketing dataset was created using the self-dialog collection method (Krause et al., 2017; Moghe et al., 2018; Byrne et al., 2019) in which a paid crowd-sourced worker writes both sides of the dialog (i.e. both customer and ticketing agent turns) based on a particular scenario…”