Information Extraction for SQL Query Generation in the Conversation-Based Interfaces to Relational Databases (C-BIRD)

Natural Language Interface to Relational Database (NLI-RDB) Through Object Relational Mapping (ORM)

“…The second approach is the automatic training approach, which does not require a knowledge engineer, instead, it only requires someone who knows well the domain, and then the task is to annotate a corpus of texts for the information being extracted. An IE method can be a part of various systems such as Knott et al [9], which uses IE to analyses financial discussion boards for automated crime detection. Owda et al [16] incorporate IE techniques into an Enhanced Conversation-Based Interface to Relational Databases (C-BIRD) in order to generate dynamic SQL queries.…”

“…Since then CA interfaces have been used effectively in many applications such as customer service, help desk, website navigation, technical support [21], web-based guidance [22], tutoring [23], assessment and training [24] and database interfaces [1,2,25]. There have been a number of CAs used as an interface to relational database such as C-BIRD [1,9], which allows a user to converse with a relational database in order to retrieve answers to queries without knowledge of SQL. The C-BIRD methodology combines pattern-matching conversational agent with knowledge trees and information extraction modules.…”

“…Object Oriented Programming (OOP) has been used as a major vehicle for developing interfaces to the RDBMSs for human friendly information retrieval and manipulation purposes. The OOP has been used to develop many NLI-RDBs [1,9,25]. However; a number of challenges have been highlighted in the literature such as impedance mismatch i.e.…”

“…NLI-RDBs research is an important area since 1960s and are still an interesting research field. In the literature, several NLI-RDB architectures have been proposed for developing prototype systems [1,2,6,7,8,9,10], which can be grouped in five main architectures: Pattern-Matching Architecture, Intermediate Representation Language Architecture, Syntax-Based Architecture, Semantic Grammar-Based Architecture and Intelligent Agents-Based Architecture [1,2,4]. The first approach is based on Pattern-Matching, which is one of the earliest architectures that used for developing NLI-RDBs.…”

Development of an Arabic Conversational Intelligent Tutoring System for Education of children with ASD

“…The current phase of the research focuses on two research questions: [12]. Also, CAs are used effectively in many application such as student's debt management guidance [13], database interfacing [14], and as student's assistant by providing problem-solving advice as they learn [15]. There are two categories of CAs, 'Embodied CAs' (ECA) and linguistic CA's (LCA).…”