Three major obstacles to the development of chat bots: the most difficult to gain user trust and understanding

The US technology blog site VentureBeat recently published an article pointing out three major obstacles in the development of chat bots, including language complexity, user trust and understanding, and ease of use. The following is the main content of this article.

Although we see chat robots as new things in the technology industry, the dialogue between people and robots has always existed. For a long time, from chatbots such as Cleverbot and Smarter Child to labyrinth phone trees, we have been working hard to create technologies that emulate our human interaction. The latest technological advances have led to the development of these tools and brought them back to the forefront of future technological dialogue.

Thanks to the virtual assistants such as Apple Siri and Amazon Alexa or the “chat bot”, the dialogue interface for voice and text has risen to a very prominent position. In addition, text-based chat bots or information platforms, such as Slack and Facebook Messenger, have made great strides in applications. These gratifying advances have also attracted a lot of investment in the industry, and have also aroused people's attention to the industry. Important brands like Domino are using virtual assistants to help customers order food, while medical companies such as Healthtap are also using Facebook Messenger to connect patients to doctors.

Although the corresponding platforms of these companies have been successful, it is important to understand that this industry will face some inevitable obstacles in the future development path. Can such a technology perform tasks according to our imagination? To what extent do users rely on chat bots? These questions reflect the user's doubts, but they are also some of the obstacles facing the technology.

For these questions, although we still have no clear answer at present, we can predict the future or problems of chat robots based on the existing knowledge, based on the historical profile of chat bots and the interactions between people and robots. . Here, we believe that chat robots will face three major obstacles in the future development path.

Modern conversational interface applications rely primarily on advances in artificial intelligence and the popularity of networked devices to provide users with shortcuts to perform simple tasks, such as answering simple questions (such as how the weather in Chicago?), or completing a Quick tasks (such as reminding me to call someone after 30 minutes) and so on.

Like software, language is based on a set of rules that evolve and evolve at any time. However, the natural person who speaks is different from the computer and is not limited by these rules. He is free to group words and sentences to convey a message. In addition to regional dialects, some individuals also develop unique linguistic methods, and humans are also good at understanding each other's language, even if the grammar is seriously deviated from the linguistic rules. When the famous American talk show host Stephen Colbert said the word "truthiness," it was not difficult to find out what he meant, even though the word was not yet in the dictionary. From a traditional perspective, computers are more difficult to understand similar vocabulary. Although machine learning has improved the language recognition ability of computers, we have not developed to the extent that artificial intelligence can catch up with the rapid evolution of language, and we cannot understand the special way of speaking in a timely manner.

In traditional conversational interfaces, comprehension is often closely tied to the user's visual or physical actions. As conversations develop in the computer field, the ability of the system to understand and respond to users' unique language approaches, whether voice input or manual input, will increase. A system that only perfectly understands structured languages ​​will be difficult to meet the needs of a large number of users, and users may find that such systems need to be improved to adapt to the computer. Otherwise, such a system that cannot respond to similar familiar languages ​​will never be able to build the familiar relationship that many product producers expect.

We have seen a large number of effective chat bots being put into the service market, such as the banking industry. Recently, when a user calls a bank to inquire about a bank account, the bank's answering agent is no longer a natural person. This is not a rare phenomenon. Through such a robot, we began to see the connection between the chat robot and the natural person, and thus brought the distance between automation and personal attention closer. Such interactions are inherently very objective and serve the artificial intelligence business well. In general, people seem to trust computers in terms of simple facts and indexes, especially in terms of data (such as the number of account assets).

When people move past objective insights into more subjective thinking, the situation becomes complicated. Telemedicine allows patients to receive medical treatment from a remote communication tool without leaving their home. This situation is becoming more and more popular in the medical field. This transformation from face-to-face to digital has been successful, mainly because at the other end of the distance, natural people are still seeing a doctor. If a user uses a robot instead of a natural person to act as a doctor to treat a patient, will the patient's trust in the robot be so high? Almost certainly not. In general, the increasing trust burden and user's doubts will become a major obstacle to the future development of chat bots.

So far, the chat bots that have been listed have not been much innovative, but there are still some advances in general compared to previous issues. For example, users can view account assets, purchase clothes, and book flowers with the help of a robot. These processes are very similar in most cases, no matter which web browser they are from. To be successful, chat bots need to provide a better experience than they currently have.

For chat bots, a better experience is simple, convenient, and seamless. It's really not fun to let the robot pick a list of books, especially if the user knows that they can pick out the books they want in a short time on the website. For some tasks, chat bots may never find a solution because they may not be able to handle these tasks simply enough. However, for other tasks, simplification, personalization, and effective machine learning can help solve many of these problems. By understanding users more thoroughly, these systems can provide unique value and streamline tasks, allowing conversations to have the simplest solution.

As time goes by, language processing and the overall level of artificial intelligence will continue to increase, and new opportunities for chat robots to perform complex tasks with less input will be opened. Of course, although most robots can follow the natural person's processing procedures, artificial intelligence still has a long way to go to solve a large number of complex natural person problems. However, when we saw significant investments such as SmarterChild and Moviefone to drive continued progress in the field, we felt that this step would be getting closer.

ES Control Cabinet

Es Control Cabinet,Customization Es Control Cabinet,Electrical Meter Control Cabinet,Outdoor Es Control Cabinet

Huaian Qiangsheng Cabinet Co., Ltd. , https://www.qscontrolcabinet.com