The AI Behind the Comments Section

2019-11-27

Recently, Zendesk discovered a new application for artificial intelligence. AI can be used to understand customer conversations, and with the help of AI, the efficiency of the company's service personnel has increased by more than ten times. This new method is highly beneficial for introducing and selling company products. Most shopping or service platforms in China have comment sections, and many people are familiar with the phrase "give a good review". Zendesk is no exception. They have a dedicated forum for customers to leave comments and suggestions. However, just like many comment sections in China, not every message is useful. Finding constructive feedback among thousands, or even millions, of message replies is undoubtedly a daunting task. From a technical perspective, the key challenge lies in the difficulty of obtaining all the text written by people on a large scale and understanding its core themes. Zendesk uses artificial intelligence to delve into these message texts and attempts to understand the emotions of the message senders using AI. The results are impressive. AI helps the company understand the common issues that people are concerned about and know what customers like and dislike. This has a significant guiding effect on their future product development. In addition, the company can directly filter out customer requests and identify common pain points among different customers. They can improve their knowledge base content and establish direct connections with customers, ensuring that they have the information customers need without having to hire information agents at critical moments. Behind this series of operations, it is actually much more complex than it seems. First, it is necessary to find an algorithm that can understand how similar two pieces of text are. This requires a deep understanding of how to effectively map each piece of text into a comparable set of numbers. What's even more difficult is determining which text fragments are similar to each other and which are different from other text fragments. Understanding the meaning of text is simple for humans, but not for machines. In simple terms, machines need to convert literary and emotional elements into mathematical and rational elements. Only in this way can they understand human text. With the help of artificial intelligence, the company has collected a lot of customer data, which helps them communicate with customers and know when to compromise or be firm without losing customers. This allows them to provide high-quality services and products to customers.

RECENT AI NEWS

RECENT AI TOOLS