Giovanni Maria Biancofiore,Yashar Deldjoo,Tommaso Di Noia,Eugenio Di Sciascio,Fedelucio Narducci
Abstract:Question answering systems are recognized as popular and frequently effective means of information seeking on the web. In such systems, information seekers can receive a concise response to their query by presenting their questions in natural language. Interactive question answering is a recently proposed and increasingly popular solution that resides at the intersection of question answering and dialogue systems. On the one hand, the user can ask questions in normal language and locate the actual response to her inquiry; on the other hand, the system can prolong the question-answering session into a dialogue if there are multiple probable replies, very few, or ambiguities in the initial request. By permitting the user to ask more questions, interactive question answering enables users to dynamically interact with the system and receive more precise results. This survey offers a detailed overview of the interactive question-answering methods that are prevalent in current literature. It begins by explaining the foundational principles of question-answering systems, hence defining new notations and taxonomies to combine all identified works inside a unified framework. The reviewed published work on interactive question-answering systems is then presented and examined in terms of its proposed methodology, evaluation approaches, and dataset/application domain. We also describe trends surrounding specific tasks and issues raised by the community, so shedding light on the future interests of scholars. Our work is further supported by a GitHub page with a synthesis of all the major topics covered in this literature study. <a class="link-external link-https" href="https://sisinflab.github.io/interactive-question-answering-systems-survey/" rel="external noopener nofollow">this https URL</a>
Computation and Language,Artificial Intelligence,Human-Computer Interaction,Information Retrieval
What problem does this paper attempt to address?
The problems that this paper attempts to solve mainly focus on the development and application of **Interactive Question Answering (IQA)**. Specifically, the paper aims to:
1. **Review existing research**: Provide a detailed review covering the popular interactive question - answering methods in the current literature. This includes explanations of the basic principles, classifications of these methods, and how to integrate various research results within a unified framework.
2. **Definition and classification**: Clearly define the implementation purposes of question - answering systems and describe in detail the main features that intelligent question - answering systems (IQASs) should possess. In addition, the paper also proposes a unified architecture, highlighting the main components of IQASs and their operation modes.
3. **Methodology and application analysis**: Conduct in - depth classification and discussion of existing IQASs from both methodological and application perspectives. This includes descriptions of specific trends in different tasks and problems raised by the community, thereby revealing the research interests of future scholars.
4. **Datasets and evaluation**: Provide detailed insights into commonly used evaluation protocols and metrics, which are carefully organized according to the paradigms and goals of IQASs.
### Key contributions of the paper
- **Formal definition**: The paper provides a formal definition of IQASs, including the question - answering problem (QA problem), two configurations of IQASs (for disambiguation and exploration), and the definition of Conversational QA Systems (CoQASs).
- **Concept of interactive session**: Introduce the concept of "Interactive Session", define the set of subsequent actions of the user after the system's response, and how these actions affect the interaction types (exploration or disambiguation).
- **Conversation state and history**: Emphasize the importance of maintaining conversation state and history in multi - round dialogues to handle complex problems such as ellipsis and ensure the coherence and relevance of the conversation.
### Specific problems solved
- **Disambiguation**: When there are multiple possible answers or ambiguity in the user's query, the system clarifies the user's intention by asking new questions to find the most relevant answer. For example, for a query such as "When was Milan founded?", the system may further ask, "Do you mean AC Milan or the city of Milan?"
- **Exploration**: After the system answers the initial query, it can suggest a series of related questions to guide the user to further explore related topics. For example, after answering "Who wrote 'The Lord of the Rings'?", the system can suggest, "Do you know the setting of 'The Lord of the Rings'?" or "What other works did Tolkien write?"
- **Conversation management**: By managing the conversation state and history, the system can handle context references and continuity problems in multi - round dialogues, ensuring the coherence and relevance of the conversation.
In conclusion, through a comprehensive analysis of existing research, this paper proposes a comprehensive framework aimed at promoting the development of interactive question - answering systems, especially in terms of disambiguation and exploration applications. This not only helps to improve the performance of the system but also enhances the user experience, making users more efficient and accurate in obtaining information.