Recently, “Watson Build Challenge” kicked off in Beijing.Synthesis Electronic Technology Co., Ltd. competed with a number of IBM partners from Hong Kong, Taiwan, Beijing, Shanghai and Guangzhou. According to judges’ comments and review on the spot, SYNTHESIS successfully ranked among top five in China area with its“Watson +Intelligent Upgrading of Bank-hospital Kiosks” solution, and will continue to participate in the next round of competition.
This Challenge is committed to promoting the competition among enterprises engaged in applications of commercial AI and pushing the implementation and development of commercial AI cases. The cases in the competition cover medical, logistics, financial, legal and other fields, and many application cases of commercial AI surprised the judges and audiences.
SYNTHESIS’s “Watson + Intelligent Upgrading of Bank-hospital Kiosks” solution has made technical breakthroughs in the following three aspects: first, in the man-machine interaction process based on natural language, there are a variety of ways to ask the same question. SYNTHESIS successfully solved this issue of synonyms with Watsons machine learning technology, avoiding the enormous workload caused by complete dependence on the configuration of synonymous dictionaries and synonym rules. Second, a number of knowledge bases are established by enumerating all questions, resulting in huge building and maintenance workload. SYNTHESIS successfully turned natural language problem into structured, parametric logical reasoning problems by virtue of Watson’s natural language analysis technology, simplifying the process ofknowledge base construction. Third, in the course of answering customer questions by means of knowledge bases, SYNTHESIS developed a multi-round dialogue mechanism to provide customers with more accurate answers and better experience. SYNTHESIS left a deep impression on the judges and audiences with its leading Watson application fruits and promising market prospects.
IBM News Link: Make AI Applicable in Reality | IBM "Watson Build Challenge"