Unlock instant, AI-driven research and patent intelligence for your innovation.

Multi-stage multi-task text reading understanding method and device

A reading comprehension and multi-task technology, applied in the field of multi-stage multi-task text reading comprehension methods and devices, can solve the problems of model effect dependence, inability to acquire effective experience, etc., so as to improve model effect, improve model accuracy, and reduce dependent effect

Pending Publication Date: 2022-01-14
SHANGHAI MININGLAMP ARTIFICIAL INTELLIGENCE GRP CO LTD
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The disadvantage of this existing technology is that the effect of the model depends on the size of the specific training data, and effective experience cannot be learned from other data or the same type of training data.

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Multi-stage multi-task text reading understanding method and device
  • Multi-stage multi-task text reading understanding method and device
  • Multi-stage multi-task text reading understanding method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0036] Preferred embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although preferred embodiments of the invention are shown in the drawings, it should be understood that the invention may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.

[0037] The terminology used in the present invention is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used herein and in the appended claims, the singular forms "a", "the", and "the" are intended to include the plural forms as well, unless the context clearly dictates otherwise. It should also be understood that the term "and / or" as used herein refers to and includes any and all possible combinations o...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention relates to a multi-stage and multi-task text reading understanding method and device. The method comprises the following steps: pre-training a text classification model by adopting a first training set, wherein the first training set is different types of questions in the same business direction; training the pre-trained text classification model by adopting a second training set, wherein the second training set is questions of the same type in the similar business direction; and predicting a to-be-predicted text by using the text classification model, and outputting a prediction result. According to the technical scheme, knowledge of other question types and knowledge of other business directions can be fused, training samples of the same business direction and answer paradigms of the same question are increased, and the model effect is further improved; for a business scene with few training samples, other data of the same business can be effectively utilized for pre-training, so that more priori knowledge is obtained, the model precision is effectively improved, the dependence on the size of a specific training data set is reduced, and the method is suitable for solving the business problem with few samples.

Description

technical field [0001] The invention relates to the field of text processing, in particular to a multi-stage and multi-task text reading comprehension method and device. Background technique [0002] In today's era of vigorous development of artificial intelligence, people are becoming more and more interested in intelligent question answering and machine reading comprehension, such as IBM Watson, Baidu Xiaodu speakers, etc., training a machine capable of text intelligence has become an important tool for various companies. A way to show strength. However, how to improve the accuracy and comprehensiveness of multiple choice reading comprehension has become an urgent problem to be solved. [0003] Aiming at this problem, an existing technology currently used is to store all standard question and answer sentences in the form of FAQ, and then when the user enters a question sentence, match the most similar question sentence, and return the result of the similar question senten...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06F40/189G06F16/35
CPCG06F40/189G06F16/35
Inventor 邓礼志于皓张杰吴信东吴明辉
Owner SHANGHAI MININGLAMP ARTIFICIAL INTELLIGENCE GRP CO LTD