In the previous post of this series, we had looked into how to design a system to match a passenger with a cab and the driver for single passenger trips (like Uber Go, or Uber Premium in India). The challenge there was matching a cab with a single passenger and matching a passenger with a single cab. But when we talk about cab pooling, we need to match a cab […]
In this series of posts we will be looking to design a cab hailing service similar to Uber or Ola (in India). We will be mainly concerned about the technical design and challenges and not get into the logistics such as signup and recruitment of drivers, training drivers for customer satisfaction, number of cabs on street and so on. Even for the technical design, we will omit some of the […]
In continuation of my earlier posts on designing a question-question similarity framework, in part three of the series we look into how to incorporate limited amount of supervised feedback into our system. Note that since getting labelled data is an expensive operation from the perspective of our company resources, the amount of supervised feedback from human agents is very low (~ 2-3% of the total number of questions). So obviously […]
In this post we will look at the offline implementation. In our company, there are currently about a 100 manual agents, each serving somewhere around 60-80 customers (non-unique) a day, i.e. a total of about 8K customer queries each day for our agents. And each customer session has an average of 3 question-answer rounds including statements, greetings, contextual and personal questions. Thus on average we generate 24K QA pairs each […]
Natural Language Question Answering system such as chatbots and AI conversational agents requires answering customer queries in an intelligent fashion. Many companies employ manual resources to answer customer queries and complaints. Apart from the high cost factor with employing people, many of the customer queries are repetitive in nature and most of the time same intents are asked in different tones.