An intelligent natural language query processor for a relational database

Moment localization in videos using natural language refers to finding the most relevant segment from videos given a natural language query. Most of the existing methods require video segment candidates for further matching with the query, which leads to extra computational costs, and they may also not locate the relevant moments under any length evaluated. To address these issues, we present a lightweight single-shot semantic matching network (SSMN) to avoid the complex computations required to match the query and the segment candidates, and the proposed SSMN can locate moments of any length theoretically. Using the proposed SSMN, video features are first uniformly sampled to a fixed number, while the query sentence features are generated and enhanced by GloVe, long-term short memory (LSTM), and soft-attention modules. Subsequently, the video features and sentence features are fed to an enhanced cross-modal attention model to mine the semantic relationships between vision and language. Finally, a score predictor and a location predictor are designed to locate the start and stop indexes of the query moment. We evaluate the proposed method on two benchmark datasets and the experimental results demonstrate that SSMN outperforms state-of-the-art methods in both precision and efficiency.

Download Full-text

Empowering Natural Language Interfaces to Databases with Aggregations

Journal of Information and Data Management ◽

10.5753/jidm.2021.1908 ◽

2021 ◽

Vol 12 (5) ◽

Author(s):

Alexandre F. Novello ◽

Marco A. Casanova

Keyword(s):

Natural Language ◽

Relational Database ◽

Natural Language Interfaces ◽

Aggregation Functions ◽

Natural Language Interface ◽

Measurement Units ◽

Database Interface ◽

Multiple Attributes ◽

Generic Module ◽

Query Aggregation

A Natural Language Interface to Database (NLIDB) refers to a database interface that translates a question asked in natural language into a structured query. Aggregation questions express aggregation functions, such as count, sum, average, minimum and maximum, and optionally a group by clause and a having clause. NLIDBs deliver good results for standard questions but usually do not deal with aggregation questions. The main contribution of this article is a generic module, called GLAMORISE (GeneraL Aggregation MOdule using a RelatIonal databaSE), that extends NLIDBs to cope with aggregation questions. GLAMORISE covers aggregations with ambiguities, timescale differences, aggregations in multiple attributes, the use of superlative adjectives, basic recognition of measurement units, and aggregations in attributes with compound names.

Download Full-text

A Natural Language Query Framework for the Semantic Web

Journal of Korean institute of intelligent systems ◽

10.5391/jkiis.2008.18.1.127 ◽

2008 ◽

Vol 18 (1) ◽

pp. 127-132

Author(s):

Jin-Sung Kim

Keyword(s):

Semantic Web ◽

Natural Language ◽

Natural Language Query

Download Full-text

Knowledge Graph Question and Answer System for Mechanical Intelligent Manufacturing Based on Deep Learning

Mathematical Problems in Engineering ◽

10.1155/2021/6627114 ◽

2021 ◽

Vol 2021 ◽

pp. 1-8

Author(s):

Miaoyuan Shi

Keyword(s):

Deep Learning ◽

Natural Language ◽

Intelligent Manufacturing ◽

Focus Of Attention ◽

Knowledge Graph ◽

Single Entity ◽

Question And Answer ◽

Natural Language Query ◽

The One ◽

Answering Questions

With the development of deep learning and its wide application in the field of natural language, the question and answer research of knowledge graph based on deep learning has gradually become the focus of attention. After that, the natural language query is converted into a structured query sentence to identify the entities and attributes in the user’s natural language query and the specified entities and attributes are used to retrieve answers to the knowledge graph. Using the advantage of deep learning in capturing sentence information, it incorporates the attention mechanism to obtain the semantic vector of the relevant attributes in the query and uses the parameter sharing mechanism to insert candidate attributes into the triple in the same model to obtain the semantic vector of typical candidates. The experiment measured that under the 100,000 RDF dataset, the single entity query of the MIQE model does not exceed 3 seconds, and the connection query does not exceed 5 seconds. Under the one-million RDF dataset, the single entity query of the MIQE model does not exceed 8 seconds, and the connection query will not be more than 10 seconds. Experimental data show that the system of knowledge-answering questions of engineering of intelligent construction based on deep learning has good horizontal scalability.

Download Full-text

Two New Challenging Resources to Evaluate Natural Language Interfaces to Databases Generated Based on Geobase and Geoquery

Handbook of Research on Natural Language Processing and Smart Service Systems - Advances in Computational Intelligence and Robotics ◽

10.4018/978-1-7998-4730-4.ch004 ◽

2021 ◽

pp. 70-100

Author(s):

Juan Javier González-Barbosa ◽

Juan Frausto Solís ◽

Juan Paulo Sánchez-Hernández ◽

Julia Patricia Sanchez-Solís

Keyword(s):

Natural Language ◽

Relational Database ◽

Normalization Procedure ◽

Natural Language Interfaces ◽

Language Problems

Databases and corpora are essential resources to evaluate the performance of Natural Language Interfaces to Databases (NLIDB). The Geobase database and the Geoquery corpus (Geoquery250 and Geoquery880) are among the most commonly used. In this chapter, the authors analyze both resources to offer two elaborate resources: 1) N-Geobase, which is a relational database, and 2) the corpus Geoquery270. The former follows the standard normalization procedure, then N-Geobase has a schema similar to enterprise databases. Geoquery270 consists of 270 queries selected from Geoquery880, preserving the same kind of natural language problems as Geoquery880, but with more challenging issues for an NLIDB than Geoquery250. To evaluate the new resources, they compared the performance of the NLIDB using Geoquery270 and Geoquery250. The results indicated that Geoquery270 was the harder corpus, while Geoquery250 is the easier one. Consequently, this chapter offers a broader range of resources to NLIDB designers.

Download Full-text