Hiring Now: A Skill-Aware Multi-Attention Model for Job Posting Generation

Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 3096–3104 July 5 - 10, 2020. c⃝2020 Association for Computational Linguistics 3096 Hiring Now: A Skill-Aware Multi-Attention Model for Job Posting Generation Liting Liu1, Jie Liu2∗, Wenzheng Zhang2, Ziming Chi2, Wenxuan Shi1, Yalou Huang1 1College of Software, Nankai University, Tianjin, China 2College of Artificial Intelligence, Nankai University, Tianjin, China {liu liting, wzzhang}@mail.nankai.edu.cn zimingchi@163.com {jliu, shiwx, huangyl}@nankai.edu.cn Abstract Writing a good job posting is a critical step in the recruiting process, but the task is often more difficult than many people think. It is challenging to specify the level of education, experience, relevant skills per the company in- formation and job description. To this end, we propose a novel task of Job Posting Gen- eration (JPG) that is cast as a conditional text generation problem to generate job require- ments according to the job descriptions. To deal with this task, we devise a data-driven global Skill-Aware Multi-Attention generation model, named SAMA. Specifically, to model the complex mapping relationships between in- put and output, we design a hierarchical de- coder that we first label the job description with multiple skills, then we generate a com- plete text guided by the skill labels. At the same time, to exploit the prior knowledge about the skills, we further construct a skill knowledge graph to capture the global prior knowledge of skills and refine the generated results. The proposed approach is evaluated on real-world job posting data. Experimental results clearly demonstrate the effectiveness of the proposed method1. 1 Introduction Writing high-quality job postings is the crucial first step to attract and filter the right talents in the re- cruiting process of human resource management. Given job descriptions and basic company informa- tion, the key to the job posting is to write job re- quirements, which requires to specify professional skills properly. Both too many or few requirements may lead to negative impacts on talent recruiting. Because of the extremely large number of job po- sitions and varieties of professional skills, a lot of ∗Corresponding Author 1https://github.com/NKU-IIPLab/SAMA Basic Information Job Description Position: Market Researcher company scale: 1000 ~ 10000 1. Assist the General Manager in sourcing travel industry news and in conducting product research and analysis. 2. Facilitate effective communication between the market research and user experience teams. 3. Translate key industry texts and compose newsletters for internal communication. Job Requirement 1. 3+ years of research experience at investment banks. 2. Strong research, data analysis and communication skills. 3. Proficient user of Microsoft Suite/G Suite. Figure 1: An example of automatic job posting. companies have to pay much cost in this step to win in the war of talents. To this end, we propose the task of Job Posting Generation (JPG) in this paper, and we cast it as a novel conditional text generation task that gen- erates the job requirement paragraph. Exploiting the ubiquitous job posting data, we aim to auto- matically specify the level of necessary skills and generate fluent job …

Content truncated to 500 words