Real estate market analysis method and device based on deep transfer learning and equipment

A technology of transfer learning and market analysis, applied in the field of real estate market analysis based on deep transfer learning, can solve problems such as time-consuming and labor-intensive, failure to achieve the effect of public opinion analysis, etc.

Active Publication Date: 2020-10-23
芽米科技(广州)有限公司
View PDF9 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the accuracy of deep neural networks depends on large-scale and high-quality labeled data. With the continuous increase of public opinion data, a large amount of manpower...

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
  • Real estate market analysis method and device based on deep transfer learning and equipment
  • Real estate market analysis method and device based on deep transfer learning and equipment
  • Real estate market analysis method and device based on deep transfer learning and equipment

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0051] This embodiment provides a specific implementation method of real estate market analysis based on deep transfer learning.

[0052] Collect real estate network public opinion data and preprocess the public opinion data. The preprocessing process here includes but is not limited to removing duplicate data, special symbols, and combining domain knowledge to mark the emotional polarity of a small amount of data to construct a real estate network public opinion dataset.

[0053] In this embodiment, the deep multi-channel neural network incorporating variational information bottleneck includes a context information extraction module, a local feature extraction module, a feature fusion module, a feature compression module and an emotion output module. Among them, the context information extraction module extracts the context information of the text through multiple Bi-GRUs; the local feature extraction module extracts local features through multiple CNNs with different sizes o...

Embodiment 2

[0079] In this embodiment, a solution method of the function max[I(Y,Z)-βI(X,Z)] is given.

[0080] In the actual calculation process, the variational inference is used to construct a lower bound for the above formula, that is, to introduce the fitted probability distribution q(y|z) and r(z) to the real probability distribution p(y|z) and p(z) Variational approximation, according to the concept that the Kullback–Leibler divergence is always positive, the final optimization goal is the variational lower bound of the original optimization goal, which can be expressed as:

[0081]

[0082] According to empirical data distribution The lower bound L can be approximated as:

[0083]

[0084] Among them, q(y|z) and q(y n |z) is the fitted conditional probability distribution, r(z) is the fitted probability distribution, p(x,y) is the real joint probability distribution, p(x) is the true probability distribution, p(y|x) and p( z|x n ) is the true conditional probability dis...

Embodiment 3

[0086] exist image 3 In the method, transfer learning is used to train the deep multi-channel neural network integrated with the variational information bottleneck, the source domain is used for pre-training, and the target domain is used for fine-tuning until the training is completed.

[0087] For the pre-training process, refer to image 3 In the first plane, the source domain data is preprocessed and input to the deep multi-channel neural network, and the deep multi-channel neural network will output its emotional features; the emotional features are fused and input into the variational information bottleneck to extract The key semantic features that affect sentiment analysis; finally output its sentiment tendency by introducing multiple fully connected layers of Maxout activation function.

[0088] For the network fine-tuning process, referring to the above pre-training process, the network weights of the context information extraction module, local feature extraction m...

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 belongs to the field of natural language processing and sentiment analysis, and particularly relates to a real estate market analysis method and device based on deep transfer learning, and equipment, and the real estate market analysis method comprises the steps: collecting real estate network public opinion data, and carrying out the preprocessing of the public opinion data; constructing a deep multi-channel neural network integrated with a variational information bottleneck; pre-training the network by using a large amount of annotation data in the related field; finely adjusting a pre-established network by using a small amount of marked public opinion data in a transfer learning mode; and performing emotional tendency analysis on unlabeled real estate public opinion dataon the migrated network, and obtaining a final real estate market emotional analysis result. According to the real estate market analysis method, deep migration learning and real estate network publicopinions are combined, and real estate market emotion can be accurately analyzed, so that reference and guidance are provided for policy making of related departments, decision deployment of companies and enterprises and investment planning of individual consumers.

Description

technical field [0001] The invention belongs to the fields of natural language processing and sentiment analysis, and in particular relates to a real estate market analysis method, device and equipment based on deep transfer learning. Background technique [0002] The growing number of mobile phone users has led to the gradual rise of social networks. Various online media and social platforms have become one of the important ways for people to obtain, disseminate and discuss social public opinion. According to the 44th "Statistical Report on Internet Development in China" issued by China Internet Network Information Center (CNNIC), as of June 2019, the number of Internet users in my country was 854 million, and the Internet penetration rate reached 61.2%. An increase of 25.98 million, the penetration rate increased by 1.6%. In addition, the report also shows that as of June 2019, the number of online news users in my country reached 686 million, an increase of 11.14 million...

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
IPC IPC(8): G06F16/33G06F16/35G06F16/951G06K9/62G06N3/04G06N3/08G06F40/289G06F40/30
CPCG06F16/951G06F16/3335G06F16/3344G06F16/35G06N3/08G06F40/289G06F40/30G06N3/048G06N3/045G06F18/241
Inventor 许国良顾桐洪岩韩茂林王铎雒江涛代朝东
Owner 芽米科技(广州)有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap