Prediction of job-seeking intention of college students based on group intelligence algorithm and its guidance for career planning

The increasingly large employment group has brought great employment pressure to society. Therefore, predicting the job-seeking intentions of college graduates and providing personalized career planning has become feasible methods to solve the structural contradiction of college employment. This paper proposes a better support vector machine model that utilizes the gray wolf optimization algorithm in the group intelligence algorithm, with the aim of establishing a prediction scheme for the job-seeking intentions of college graduates using this model. The collected data are pre-processed by cleaning, integrating, converting, and other pre-processing work to form a standardized dataset to prepare for the subsequent experiments. Finally, the prediction scheme of the job-seeking intention of college graduates is designed and developed based on the GWO-SVM optimization model, and the actual application of the prediction model is analyzed as a case study based on the real data of graduates from a college. The study shows that when the number of input features is 9, the average value of the comprehensive score of this prediction model is 0.5703, the standard deviation is 0.0278, the optimal learning rate of the model is 0.165, the F1 value is stable above 0.6, and the optimal F1 value is 0.856, which proves that this prediction model has a good prediction effect. This study has certain guiding significance in helping college students establish reasonable job-hunting intentions, proposing personalized career planning for students, and promoting high employability. © 2024 Guoshi Yu et al., published by Sciendo.