A Survey of Information Extraction Based on Deep Learning

Abstract

As a core task and an important link in the fields of natural language understanding and information retrieval, information extraction (IE) can structure and semanticize unstructured multi-modal information. In recent years, deep learning (DL) has attracted considerable research attention to IE tasks. Deep learning-based entity relation extraction techniques have gradually surpassed traditional feature- and kernel-function-based methods in terms of the depth of feature extraction and model accuracy. In this paper, we explain the basic concepts of IE and DL, primarily expounding on the research progress and achievements of DL technologies in the field of IE. At the level of IE tasks, it is expounded from entity relationship extraction, event extraction, and multi-modal information extraction three aspects, and creates a comparative analysis of various extraction techniques. We also summarize the prospects and development trends in DL in the field of IE as well as difficulties requiring further study. It is believed that research can be carried out in the direction of multi-model and multi-task joint extraction, information extraction based on knowledge enhancement, and information fusion based on multi-modal at the method level. At the model level, further research should be carried out in the aspects of strengthening theoretical research, model lightweight, and improving model generalization ability.

Optimal feature extraction from multidimensional remote sensing data for orchard identification based on deep learning methods

“…66,67 Meanwhile, deep learning in information mining and data processing is also showing strong abilities. 68 However, the traditional approach to processing multi or high-dimensional remote sensing data, where dimension reduction and classification are treated as separate steps, does not ensure that the final classification results will necessarily benefit from the dimension reduction process. Therefore, two new neural networks (i.e., 1D-FSA-CNN and 3D-FSA-CNN) are proposed in this paper, which combine dimension reduction and classification for orchard classification from multidimensional remote sensing data.…”

“…It is proved in some studies that multi-dimensional and multi-source remote sensing data can effectively leverage the spectral information of crops, mitigate the impact of adverse weather conditions such as clouds and rain, and facilitate the extraction of crop planting structures 66 , 67 . Meanwhile, deep learning in information mining and data processing is also showing strong abilities 68 . However, the traditional approach to processing multi or high-dimensional remote sensing data, where dimension reduction and classification are treated as separate steps, does not ensure that the final classification results will necessarily benefit from the dimension reduction process.…”

Methods of knowledge extraction from natural language texts

“…Под информацией понимаются некоторые сведения и факты в рамках определенного контекста [1]. В рамках задачи интеллектуального анализа текста, извлечение информации -это процесс распознавания в тексте некоторых сущностей, отношений, событий или другой фактографической информации, то есть получение из текста структурированных данных, пригодных для включения в целевую систему знаний и машинной интерпретации [2].…”

“…Методы на основе МО. Обзор методов извлечения отношений на основе МО можно найти во многих работах, например в [2,33,42,43]. Выигрышной стороной МО является универсальность используемого алгоритма с точки зрения вида извлекаемых отношений, проблема лишь в формировании достаточно большого корпуса так или иначе размеченных текстов и выборе набора признаков, эффективно индицирующих наличие искомого отношения между понятиями.…”

A Levitated Control Attention for Named Entity Recognition

“…It suffers from a conflicting problem: recognizing a named entity heavily depends on contextual features of a sentence, but these features are shared by different named entities at the same time. Because the NER's performance significantly influences downstream tasks, such as entity relation extraction [4], [5], event extraction [6], [7], and syntactic analysis [8], [9]. Therefore, the progress of NER has important theoretical and practical values in the field of natural language processing.…”