A semantic architecture for artificial conversations

Abstract

Abstract-Artificial conversations have many applications in chatter bot-based customer service including website navigation tools and guided online shopping. Existing approaches to generating conversations leverage linguistic and stochastic principles, where lower level grammatical and structural artifacts are modeled. These approaches perform well in pairwise utterance exchanges, but not so well in longer conversational contexts. We simulate more meaningful chatter bot conversations using an architecture that can leverage content and context. Grice's cooperative maxims, which form the central idea in the theory of pragmatics, is our framework for evaluation. The domain of our research is customer service situations, and we compare our artificial conversations with actual conversations of existing chatter bots deployed in the same domain.

Artificial conversations for customer service chatter bots: Architecture, algorithms, and evaluation metrics

“…Contemporary chatter bots perform very effectively in question-answer settings and similar utterance-exchange pair settings, where the contexts of the conversations are independent from one exchange to the next (Mauldin, 1994;Saygin & Ciceklib, 2002;Chakrabarti & Luger, 2012). However, they perform poorly in conversational situations where a specific context is maintained through a series of several utterance-exchange pairs.…”

“…However, they perform poorly in conversational situations where a specific context is maintained through a series of several utterance-exchange pairs. Existing customer service chatter bots are able to handle FAQ-type queries, but are unable to handle contexts that require a short conversation (Chakrabarti & Luger, 2012;Chakrabarti, 2014). Most chatter bot implementations focus on either content modeling or conversation semantics, or sub-aspects of these, or incorporate both of them together without making an explicit distinction.…”

“…They are data structures that seek responses to specific queries (or sub-contexts) requiring specific responses to make the conversation progress towards a goal. While engaging in a dialogue with the user, the chatter bot captures specific pieces of information from the conversation to progress along the network of contexts described by the goal-fulfillment map (Chakrabarti & Luger, 2012). A context, uniquely identified by the topic and the type of conversation, is implemented as a list of goal-fulfillment maps.…”

Chatbots for troubleshooting: A survey

“…Chatbots driven by rules are common in the troubleshooting domain (Boye, 2007;Chakrabarti & Luger, 2012) and beyond (Banks, Jiang, Kim, Niswar, & Hui Yeo, 2013;Forbell et al, 2013;Janarthanam et al, 2013;Sansonnet, Correa, Jacques, Braufort, & Verrechia, 2012). Rule-based dialog management is depicted by Figure 4.…”

Arabic Chatbots: A Survey

“…There are generally three components of Chabot [5]: the interface, which interacts with user's input and output, the Knowledge Base or brain, which include the content of the conversation and keep truck of the domain, and the Conversation Engine, which manages the semantic context of the conversation.…”