The invention relates to a C2C closed-loop job-hunting and recruitment matching and transaction guarantee platform based on a content
label algorithm, and belongs to the technical field of internet recruitment. The core innovation points comprise: 1, a decentralized matching mechanism; 2, dynamic interest tags are generated from user contents and behaviors through NLP and
reinforcement learning (DQN); and 3, forbidding the paid promotion / VIP authority, and realizing fair matching only based on the
label similarity and the historical evaluation weight. 4, a block chain enabled transaction
closed loop; 5, a labor order model is created for the first time: two parties negotiate for pricing, a block chain labor contract is signed, and a recruiter pre-stores a period salary to a
bank escrow account; and 6, after the labor cycle is finished, dynamically
settling the salary according to the actual performance data (the platform automatically transfers the actual salary and returns the remaining money). 7, the evaluation-flow linkage feedback
system performs bidirectional mutual evaluation on the data
encryption uplink, and the evaluation weight is directly associated with the subsequent
exposure priority of the user; and when the difference evaluation rate exceeds a threshold value, automatically triggering the account to drop the right or freeze. The technical effects are as follows: (1) the matching fairness is improved (test shows that the Gini coefficient is reduced by 65%); (2) the salary dispute rate is reduced to 1.5% or below; and (3) the commercialization whole-
process supervision of the talent market is realized.