In contemporary society, the problem of information asymmetry in talent markets has been becoming more prominent. On one hand, the company and candidates fight against each other based on the information available, so...In contemporary society, the problem of information asymmetry in talent markets has been becoming more prominent. On one hand, the company and candidates fight against each other based on the information available, so both of them could make fraud that will make the market level lower and lower. On the other hand, former scholars have studied from enterprises＇ perspective and put forward methods to solve it based on the aspect of improving the technology and standard mechanism, which could not solve the problem of information asymmetry thoroughly. Consequently, this research put up with the idea that the market can reduce information asymmetry through the establishing personnel information database and related platforms, which has a great practical significance on realizing the optimal allocation of the market and saving cost. At the same time, this study discussed the problems of information asymmetry fundamentally, which was of great importance to enrich the related theory research. Specific models were constructed through two perspectives from the enterprise and the candidates. And then two models would be eventually integrated into a large system. Finally, this research put all related information into a system, which was beneficial to the optimal allocation of human resources with constraints of the market environment.展开更多
Support vector machine （SVM） is an important classi- fication tool in the pattern recognition and machine learning community, but its training is a time-consuming process. To deal with this problem, we propose a nov...Support vector machine （SVM） is an important classi- fication tool in the pattern recognition and machine learning community, but its training is a time-consuming process. To deal with this problem, we propose a novel method to mine the useful information about classification hidden in the training sample for improving the training algorithm, and every training point is as- signed to a value that represents the classification information, respectively, where training points with the higher values are cho- sen as candidate support vectors for SVM training. The classifica- tion information value for a training point is computed based on the classification accuracy of an appropriate hyperplane for the training sample, where the hyperplane goes through the mapped target of the training point in feature space defined by a kernel fimction. Experimental results on various benchmark datasets show the effectiveness of our algorithm.展开更多
文摘In contemporary society, the problem of information asymmetry in talent markets has been becoming more prominent. On one hand, the company and candidates fight against each other based on the information available, so both of them could make fraud that will make the market level lower and lower. On the other hand, former scholars have studied from enterprises＇ perspective and put forward methods to solve it based on the aspect of improving the technology and standard mechanism, which could not solve the problem of information asymmetry thoroughly. Consequently, this research put up with the idea that the market can reduce information asymmetry through the establishing personnel information database and related platforms, which has a great practical significance on realizing the optimal allocation of the market and saving cost. At the same time, this study discussed the problems of information asymmetry fundamentally, which was of great importance to enrich the related theory research. Specific models were constructed through two perspectives from the enterprise and the candidates. And then two models would be eventually integrated into a large system. Finally, this research put all related information into a system, which was beneficial to the optimal allocation of human resources with constraints of the market environment.
基金Supported by the National Natural Science Foundation of China (61070137,60933009)the Science and Technology Research Development Program in Shaanxi Province of China (2009K01-56)
文摘Support vector machine （SVM） is an important classi- fication tool in the pattern recognition and machine learning community, but its training is a time-consuming process. To deal with this problem, we propose a novel method to mine the useful information about classification hidden in the training sample for improving the training algorithm, and every training point is as- signed to a value that represents the classification information, respectively, where training points with the higher values are cho- sen as candidate support vectors for SVM training. The classifica- tion information value for a training point is computed based on the classification accuracy of an appropriate hyperplane for the training sample, where the hyperplane goes through the mapped target of the training point in feature space defined by a kernel fimction. Experimental results on various benchmark datasets show the effectiveness of our algorithm.
