AI support, more efficient and accurate data usage!

Based on the Hadoop big data platform created with 5 billion trade data, the Spark computing engine and XGBOOST intelligent algorithm were introduced. Combined with the past companies’ trading products, cycles, quantity and proportion, after multiple rounds of large model training and autonomous learning, the companies that will purchase and supply related products in the next 1-2 months were predicted with an accuracy of up to 92%.

Use Hive data warehouse. After the monthly data is updated, the companies that were previously predicted to purchase and supply products are verified, and the transaction quantity of specific products and the transaction proportion of the predicted company are listed, and proof marks are made.

Multi-condition joint search, multi-product keywords, multi-HS customs codes, and multi-port joint search can meet users' joint search needs for multiple complex conditions and find more potential target customers. This can be done through advanced search.

Based on the HS code and product description, each transaction record is labeled with a product label (in English). This can solve the problem that English keywords cannot be used to search in countries with small languages; it can also filter out companies with a high product share in the past two years.

By keywords and HS codes search, directly display foreign trading partners (buyers or suppliers) that have transactions with domestic counterparts. Trading partners are sorted according to the number of transactions between the two parties, which is more efficient and accurate.

The employee contact info of the recommended companies comes from the business interface, including employee portraits, names, positions, email addresses, social media addresses, etc; it supports searching employee positions, finding key persons (KP) of the company, and supporting the export of employee contact info.

Based on the initial product requirements and subsequent search and browsing behaviors, target customers with company employee contact info are automatically recommended every day. This saves users the time cost of searching, verifying and collecting contact info, and is more efficient.