User attribute prediction method and related device based on deep learning
A technology of user attributes and prediction methods, applied in the direction of neural learning methods, neural architecture, biological neural network models, etc., can solve the problem that personal introduction information is not necessarily accurate, and achieve the effect of high accuracy
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0033] see figure 1 , figure 1 It is a schematic flowchart of an embodiment of the user attribute prediction method of the present invention. like figure 1 As shown, the process includes the following steps:
[0034] S11: Input the original text information into the first neural network model, and obtain the text sequence feature of the global feature;
[0035] In step S11, the first neural network model is a bidirectional long-short-term memory network model.
[0036] When the bidirectional long-short-term memory network model is trained, the input text information is divided into multiple words, and each word is represented by a pre-trained word vector model, and the word vector model matrix uses the Skip-Gram model;
[0037] The loss function used in the word vector model training is: Among them, V is the corpus, |V| is the length of the corpus, W t+i Indicates the floating window Indicates the context word under, C is the size of the window, f(W t+i ,W t )=P(W ...
Embodiment 2
[0060] see Figure 4 , Figure 4 It is a schematic structural diagram of an embodiment of the user attribute prediction device of the present invention. Figure 4 The device includes a character sequence feature acquisition module 41, a word block feature acquisition module 42, a subtask feature acquisition module 43 and a user attribute prediction module.
[0061] The text sequence feature acquisition module 41 is configured to input the original text information into the first neural network model to acquire the text sequence feature of the global feature.
[0062] The word block feature acquisition module 42 is configured to input the original text information into the second neural network model to acquire the word block feature of the global feature.
[0063] The subtask feature acquisition module 43 is configured to input the original text information into the third neural network model to acquire multiple subtask features.
[0064] A user attribute prediction module ...
Embodiment 3
[0067] The present invention also provides a device for predicting user attributes based on deep learning, which includes: at least one processor; and a memory connected to the at least one processor in communication; An instruction to be executed, the instruction is executed by the at least one processor, so that the at least one processor can execute the method as described in the first embodiment.
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More