Data Authenticity Analysis for Online O2O Data: A Case Study of Second-Hand Houses Posting Data

The rise of Electronic Commerce Industry makes O2O become one development tendency. As the largest classified information website, 58 city website contains most information in O2O area. However, the network environment is not as simple as we thought, fraud is ubiquitous in the market of intermediaries and information networks. There are vast amounts of information in 58, so that people have to firstly extract the information, and then make profitable decisions. This paper crawls Lianjia and 58 website data, uses BP neural network to predict the best quotation and transaction price of second-hand housing, provides the basis for the seller and intermediary to quote, and provides the transaction price expectation for the buyer and seller, so as to improve the market efficiency of the second-hand housing online trading platform. To measure the validity of the model, we propose three indicators: the Seller Market Efficiency (SEM), the Intermediary Market Efficiency (IME) and the Buyer’s Market Efficiency (BME). The results show that the model can effectively improve market efficiency.