Does your business face the following problems:
- The company plans to develop a new business. How can the CEO quickly find the right business leader through label features among the existing cadres and potential talents?
- The company has won the bid for a large project and needs to quickly form a team to deliver it. The time is tight and the tasks are heavy. How can HR find people with relevant project experience in the company through the key information of the project, and recommend them to the project team in time to ensure the normal operation of the project? conduct?
- When the boss of the company finds an excellent employee, he tells HR, how many excellent employees like "him" are there in our company? How can HR find talents with similar characteristics through the key characteristics of excellent employees?
- When the company conducts internal rotation of cadres and talents, the boss wants to know which cadres are suitable for this position. How can HR find the list of matching personnel through the job portrait standard, and can also sort them according to the matching score?
Yonyou BIP intelligent talent discovery helps users find the required talents more easily. Through multi-scenario models, such as finding people by label, finding people by position, finding people by person, and finding people by project, etc., we can match talent needs in different scenarios. For example, in models such as person-post matching and personnel similarity matching, advanced technologies such as knowledge graphs, NLP, and intent recognition are integrated to improve accuracy and matching, thereby solving the problem of talent shortage faced by users.
1. Search in the search box: support multi-keyword combination query, short text natural language input
The search box not only supports name, age, rank, position, sequence, performance, graduate school, education, label, etc., but also supports multi-keyword combination query, and can automatically understand the "and" between keywords according to the user input content or" relationship to help users find suitable talents to the greatest extent.
The short text natural language input method is more in line with user habits, and the corresponding search intent is identified through NLP analysis, which greatly improves the user experience.
2. Tags to find people
We display the tags commonly used by users to users in the most intuitive way, helping users to quickly find multi-dimensional combinations by clicking and find the talents they belong to.
3. Find someone by post
Connect with job portraits, find employees suitable for the current position through the person-post matching model, and recommend according to the degree of matching, support users to view the details of employee portraits, and compare and analyze multiple employees to find the most suitable talents .
4. Find people by people
Users can use an excellent employee as an input condition, and quickly find employees similar to the employee through the person-to-person model. The person-to-person model calculates the similarity of portraits to maximize the excavation of more outstanding talents.
5. Find people for the project
By selecting the project and the corresponding position, the user can match the talent who is most suitable for the corresponding position of the current project with the help of the project recruitment model. On the basis of the most suitable job model, attributes such as project category and project scale are added to make the talents found more suitable for the current project requirements.
Yonyou BIP intelligent talent discovery applies advanced technical architecture, and uses Query analysis to segment search content; uses synonym models and association word models to accurately identify keywords; recalls through text semantic correlation, ES retrieval, and combines XGB and GBT The model performs feature fusion; the search results are sorted according to the relevance of the fine-ranking model and the mixed-ranking model; the label mining model and batch marking tools enrich employee portraits and greatly improve the accuracy of talent matching; finally, use User behavior and employee attribute data, combined with the intelligent recommendation model, makes the push results more in line with user expectations, and realizes the personalization and precision of talent recommendation.