Tag: NLP

Novelis technical experts have once again achieved a new state-of-the-art in science. Discover our study SQL Generation from Natural Language: A Sequence-to-Sequence Model Powered by the Transformers Architecture and Association Rules, puplished on Journal of Computer Science.

Thanks to the Novelis Research Team for their knowledge and expertise.

Abstract

Using natural language (NL) to interact with relational databases allows users of any background to easily query and analyze large amounts of data. This requires a system that understands user questions and automatically translates them into structured query languages ​​(such as SQL). The best-performing Text-to-SQL system uses supervised learning (usually expressed as a classification problem) and treats this task as a sketch-based slot filling problem, or first converts the problem into an intermediate logical form (ILF) and then converts it Convert to the corresponding SQL query. However, unsupervised modeling that directly translates the problem into SQL queries has proven to be more difficult. In this sense, we propose a method to directly convert NL questions into SQL statements.

In this research, we propose a sequence-to-sequence (Seq2Seq) parsing model for NL to SQL tasks, supported by a converter architecture that explores two language models (LM): text-to-text transfer converter (T5) ) And multi-language pre-trained text-to-text converter (mT5). In addition, we use transformation-based learning algorithms to update aggregation predictions based on association rules. The resulting model implements a new state-of-the-art technology on the WikiSQL data set for weakly supervised SQL generation.

About the study

“In this study, we treat the Text-to-SQL task with WikiSQL1 (Zhong et al., 2017). This DataSet is the first large-scale dataset for Text-to-SQL, with about 80 K human-annotated pairs of Natural Language question and SQL query. WikiSQL is very challenging because tables and questions are very diverse. This DataSet contains about 24K different tables.

There are two leaderboards for the WikiSQL challenge: Weakly supervised (without using logical form during training) and supervised (with logical form during training). On the supervised challenge, there are two results: Those with Execution Guided (EG) inference and those without EG inference.”

Read the full article

Journal of Computer Science – Volume 17 No. 5, 2021, 480-489 (10 pages)

Journal of Computer Science aims to publish research articles on the theoretical basis of information and computing, and practical technologies for implementation and application in computer systems.

Discover our conference paper SQL Generation from Natural Language Using Supervised Learning and Recurrent Neural Networks – International Conference on Artificial Intelligence & Industrial Applications part of the Lecture Notes in Networks and Systems book series (LNNS, volume 144).

Thanks to the Novelis Research Team for their knowledge and expertise.

Abstract

The database stores today’s large amounts of data and information. To access these data, users need to master SQL or an equivalent interface language. Therefore, using a system that can convert natural language into equivalent SQL queries will make the data more accessible. In this sense, building a natural language interface to a relational database is an important and challenging problem in the field of natural language processing (NLP) and extensive research, and due to the introduction of large-scale data sets, it has recently been discovered again momentum. In this article, we propose a method based on word embedding and recurrent neural network (RNN), precisely based on long short-term memory (LSTM) and gated recurrent unit (GRU) units. We also showed the dataset used to train and test our model, based on WikiSQL, and finally we showed our progress in accuracy.

About the study

“Vast amount of today’s information is stored in relational database and provide the foundation of applications such as medical records [1], financial markets [2], and cus- tomer relations management [3]. However, accessing relational databases requires an understanding of query languages such as SQL, which, while powerful, is difficult to master for non-technical users. Even for an expert, writing SQL queries can be chal- lenging, as it requires knowing the exact schema of the database and the roles of various entities in the query. Hence, researches has recently appeared to approach systems that map natural language to SQL query, and a long-standing goal has been to allow users to interact with the database through natural language [4,5]. We refer to this task as Text-to-SQL.

In this work, we present our approach based on Classifications [6] and Recurrent Neural Networks [7], precisely on LSTM [8] and GRU [9] cells. The idea is inspired from SQLNet approach [10]; in particular, we employ a sketch to generate a SQL query from naturel language. The sketch aligns naturally to the syntactical structure of a SQL query; Neural Networks are then used to predict the content for each slot in the sketch. Our approach can be viewed as a neural network alternative to the traditional sketch based program synthesis approaches [11,12].”

Read the full article

Part of the Lecture Notes in Networks and Systems book series (LNNS, volume 144) 

SpringerLink provid researchers with access to millions of scientific documents from journals, books, series, protocols, reference works and proceedings.