US 12,086,556 B2
Method, apparatus, device and medium for generating recruitment position description text
Chuan Qin, Beijing (CN); Kaichun Yao, Beijing (CN); Hengshu Zhu, Beijing (CN); Chao Ma, Beijing (CN); Dazhong Shen, Beijing (CN); and Tong Xu, Beijing (CN)
Assigned to Baidu Online Network Technology (Beijing) Co., Ltd., Beijing (CN)
Filed by Baidu Online Network Technology (Beijing) Co., Ltd., Beijing (CN)
Filed on Mar. 26, 2021, as Appl. No. 17/214,080.
Claims priority of application No. 202010381686.9 (CN), filed on May 8, 2020.
Prior Publication US 2021/0216726 A1, Jul. 15, 2021
Int. Cl. G06F 17/00 (2019.01); G06F 40/279 (2020.01); G06F 40/30 (2020.01); G06F 40/40 (2020.01); G06N 3/08 (2023.01); G06Q 10/0631 (2023.01)
CPC G06F 40/40 (2020.01) [G06F 40/279 (2020.01); G06F 40/30 (2020.01); G06N 3/08 (2013.01); G06Q 10/063112 (2013.01)] 16 Claims
OG exemplary drawing
 
1. A method for generating a recruitment position description text, comprising:
obtaining an original text related to a target position; and
generating a target recruitment position description text corresponding to the target position based on the obtained original text and a pre-trained deep neural network model, wherein generating the target recruitment position description text corresponding to the target position based on the obtained original text and the pre-trained deep neural network model comprises extracting skill requirements of a job from the original text using the pre-trained deep neural network model,
wherein the deep neural network model comprises:
a text subject predicting sub-model for predicting a target skill subject distribution vector based on the original text; and
a description text generating sub-model for generating the target recruitment position description text of the target position based on the predicted target skill subject distribution vector; and
wherein a training process of the deep neural network model comprises:
obtaining a first training sample data, wherein the first training sample data comprises a first sample-related text of a first sample position and a first standard recruitment position description text corresponding to the first sample position;
using the obtained first training sample data to preliminarily train a pre-constructed text subject predicting sub-model to obtain a preliminary trained text subject predicting sub-model;
obtaining a second training sample data, wherein the second training sample data comprises a second sample related text of a second sample position and a second standard recruitment position description text corresponding to the second sample position; and
using the obtained second training sample data to train the deep neural network model including the preliminary trained text subject predicting sub-model and a pre-constructed description text generating sub-model, to obtain a trained deep neural network model.