Natural Language Interfaces to Data: Introduction 2. Background 3. Natural Language Querying Architectures 4. Conversational Data Analysis and Exploration 5. Benchmarks and Evaluation Techniques 6. Open Challenges 7. Conclusion References

Download or read book Natural Language Interfaces to Data: Introduction 2. Background 3. Natural Language Querying Architectures 4. Conversational Data Analysis and Exploration 5. Benchmarks and Evaluation Techniques 6. Open Challenges 7. Conclusion References written by Abdul Quamar and published by . This book was released on 2022 with total page 108 pages. Available in PDF, EPUB and Kindle. Book excerpt: Natural language interfaces provide an easy way to query and interact with data and enable non-technical users to investigate data sets without the need to know a query language. Recent advances in natural language understanding and processing have resulted in a renewed interest in natural language interfaces to data. The main challenges in natural language querying are identifying the entities involved in the user utterance, connecting the different entities in a meaningful way over the underlying data source to interpret user intents, and generating a structured query. There are two main approaches in the literature for interpreting a user's natural language query. The first are rule-based systems that make use of semantic indices, ontologies, and knowledge graphs to identify the entities in the query, understand the intended relationships between those entities, and utilize grammars to generate the target queries. Second are hybrid approaches that utilize both rule-based techniques as well as deep learning models. Conversational interfaces are the next natural step to one-shot natural language querying by exploiting query context between multiple turns of conversation for disambiguation. In this monograph, the authors review the rule-based and hybrid technologies that are used in natural language interfaces and survey the different approaches to natural language querying. They also describe conversational interfaces for data analytics and discuss several benchmarks used for natural language querying research and evaluation. The monograph concludes with discussion on challenges that need to be addressed before these systems can be widely adopted.