The first aspect: natural language processing
Natural language processing is a science that integrates linguistics, computer science and mathematics. Natural language processing does not usually study natural language, but aims at developing a computer system that can effectively realize natural language communication, especially the software system, which is a field in which computer science, artificial intelligence and linguistics pay attention to the interaction between computers and human (natural) languages. The purpose of natural language processing is to realize various theories and methods of effective communication between people and computers in natural language. safe
First, multilingual translation. machine learning
One of the main applications of natural language processing is foreign language translation. When you meet a foreign language article in your life, the first thing you think of is to find a translated web page or APP. However, every time the result of machine translation is basically not in line with the language logic, we need to input the sentence again for secondary processing and arrangement. As for translation in professional fields, such as law and medical treatment, machine translation is simply not feasible. study
Faced with this dilemma, natural language processing is trying to break through the barriers of translation. As long as it provides massive data, the machine can learn any language by itself. It takes about 2 weeks for the machine to enter a field (zero cost entry) from 0. Therefore, entering any field can be highly vertical. For example, in the translation of legal professional articles, the total amount of high-quality legal articles is limited. If you let the machine learn these articles once, you can ensure 95% fluency in translation and achieve real-time synchronization. test
Second, virtual my assistant. big data
Virtual my assistant refers to the way that users complete some small things in ordinary life through voice control and text input. Most virtual assistants can collect simple life information and help you optimize information and make intelligent decisions while watching relevant comments. optimize
At the same time, some virtual assistants can also play music directly or receive emails, which are one of the variations of virtual assistants. Virtual my assistant is applied in all aspects of our lives, such as audio, car, smart home, smart car and intelligent customer service. Generally speaking, people who can complete the service after hearing voice instructions are basically virtual assistants. cloud computing
Third, intelligent case handling artificial intelligence
Natural language processing can also automatically convert the backlog of cases into a structured database in batches. Machine learning and natural language processing technology can automatically capture clinical variables in medical records and generate a standardized database. Subsequently, the whole process of auxiliary intelligent algorithm from variable extraction, idea generation to paper chart export can mine the correlation of variables, stimulate the ideas of papers, and provide professional statistical analysis support for clinical research at the same time.
Its level is as high as that of medical graduate students who have received 8 years' clinical medical education. In this way, they can read a 50-page medical record at the same time, and the speed of grasping and understanding all the clinical information in it is 2700 times faster than that of medical life, which greatly improves the office efficiency of the hospital, and the problem of seeking medical treatment will be greatly alleviated.
The second aspect: speech recognition
Speech recognition is an interdisciplinary subject. The fields involved in speech recognition technology include: signal processing, pattern recognition, probability theory and information theory, vocal mechanism and auditory mechanism, artificial intelligence and so on. Voice communication with machines, so that machines can understand what you say, is something that people have been longing for for for a long time. Now artificial intelligence has turned this ideal into reality and brought it into our ordinary life.
First, intelligent hospitals.
Relying on artificial intelligence technology and big data, the hospital can realize intelligent voice interactive knowledge question and answer and medical record inquiry, and voice input can replace typing, so that you can easily input with computers, tablet computers and mobile ward round equipment by speaking. Everyone's words will be transcribed into words and displayed in your HIS system, PACS system, CIS system and other places where you want to input words. In addition, it can also predict health risks and analyze patient groups.
Second, oral evaluation.
There is also an interesting application in speech recognition-speech evaluation service, which uses cloud computing technology to put automatic oral evaluation service in the cloud and open API interface for customers to use remotely. In the phonetic evaluation service, interactive teaching between human and computer can realize one-on-one oral counseling, which is like inviting a foreign teacher at home and solving the problem of dumb English in the future.
The third aspect: computer vision
Computer vision is a science that studies how to make machines "see". Further, it refers to using cameras and computers to identify, track and measure targets instead of human eyes, and further processing graphics to make computers process images that are more suitable for human eyes to observe or send to instruments for detection. After computer vision, the computer will process images that are more suitable for human eyes to observe or send to instruments for detection. The main task of computer vision is to get the three-dimensional information of the corresponding scene by processing the collected pictures or videos.
First, intelligent security.
With the process of governments at all levels vigorously promoting the construction of "safe city", there are more and more monitoring points, and video and bayonet have produced a huge amount of data. Especially with the popularity of high-definition monitoring, the amount of data in the whole security monitoring field is increasing explosively, and it is becoming more and more difficult to analyze and process this information manually. The security technology field with computer vision as the core has a large number of data sources and rich data levels, and the essential demands of security business are highly consistent with the technical logic of AI, from being able to engage in preventive application before to tracing afterwards.
Second, face recognition abduction.
At present, the crime of abducting and selling children is rampant in China, with tens of thousands of victims and families. According to the estimation of the Ministry of Civil Affairs, at present, the number of vagrants and beggars in China is around100000-1500000. In rural areas such as Henan, Yunnan and the coastal areas of Guangdong and Guangxi, the sale of children is almost market-oriented, resulting in a complete underground black interest chain. It is urgent to find the abducted children. At present, the application of computer vision in "portrait recognition and face comparison" can make the abducted children be found within 7 hours at the earliest. This is a huge application of computer vision in the security field, and it will be more and more used in combating crime.
The fourth aspect: expert system.
Expert system is the most important and active application field in artificial intelligence. It refers to an intelligent computer program system that contains a lot of expert knowledge and experience in a certain field and uses human experts' knowledge and problem-solving methods to deal with problems in this field. Generally speaking, based on the knowledge and experience provided by one or more experts in a certain field, reasoning and judgment are carried out to simulate the decision-making process of human experts to solve those complex problems that need to be dealt with by human experts.
First, unmanned cars.
Driverless car is a kind of intelligent car, also known as wheeled mobile robot, which mainly relies on the intelligent driving instrument with computer system in the car to achieve the goal of unmanned driving. Since the 1970s, the United States, Britain, Germany and other developed countries have begun to study driverless cars, and made breakthrough progress in feasibility and practicality.
China began to study driverless cars in 1980s, and the University of National Defense Technology successfully developed the first truly driverless car in China in 1992. In 2005, the first urban driverless car was successfully developed in Shanghai Jiaotong University. The world's earliest driverless car has been tested and driven for nearly 500 thousand kilometers, of which the last 80 thousand kilometers were completed without any human safety intervention measures.
Second, the weather forecast
With the popularity of mobile phones, more and more people are used to watching the weather forecast in mobile phones, and the position of expert system is also decisive in the weather forecast. The expert system can first locate the user's position through the GPRS system of the mobile phone, and then use the algorithm to analyze and predict the radar map covering the whole country.
Users can query the weather trend of their own place anytime and anywhere. In the weather forecast, there is no longer the word "rain in some areas". Instead, it is "light rain in your street after 25 minutes, and it will stop after 50 minutes". Assign a dedicated weatherman to you, so that the weather forecast you receive can be accurate to the minute and the street you are in.
Third, the urban system
Urban system is to digitize all the infrastructure such as transportation, energy and water supply, gather the data scattered in all corners of the city, and then realize the global real-time analysis of the whole city through super-strong analysis and super-large-scale calculation, so that the city can run intelligently. The first problem solved by urban system is traffic jam. This year, the urban brain of Hangzhou intelligently adjusted the traffic lights by analyzing the map data and camera data, and successfully increased the speed of vehicles by 1 1%, greatly improving the travel experience.
The fifth aspect: cross-use in various fields
In fact, the four applications of artificial intelligence actually involve other fields more or less, but the most prominent aspect of cross-application is still intelligent robot. A robot is a mechanical device that automatically performs work. It can not only accept human command, but also run pre-programmed programs and act according to the principle program formulated by artificial intelligence technology. Its task is to assist or replace human work, such as production, construction, or dangerous work.
First, the logistics robot
Logistics robot is the leader of the technological change that combines robot products and artificial intelligence technology to realize highly flexible and intelligent logistics automation. Under the market pressure of consumption upgrading, the inventory management of massive SKUs and uncontrollable labor costs have all become the same problems in e-commerce, retail and other industries. The logistics robot has low management cost and strong package integrity, which can meet the requirements of various sorting efficiency and accuracy, and the return on investment is short. Its appearance can effectively improve the flexibility of production and help enterprises realize intelligent transformation, and it will be more and more used in daily life.
Second, Meng Chong robot
Children have always been the heart and soul of parents, and how to make children win at the starting line is also a matter of great concern to parents of all walks of life. At this time, early education is particularly important. In fact, early education is to let children play effectively, so that children can learn a lot of knowledge in the process of playing, develop their brain power, hands-on ability, reaction ability, aesthetic ability, and cultivate their interests and habits.
Early education institutions in the market are expensive, lack of teachers, and there may be certain security risks. At this time, the existence of cute pet robots has greatly alleviated this problem. The voice function allows it to communicate with children just like children's friends, and the memory function can also remember the baby's usage habits and quickly find what the baby wants to hear. At the same time, it provides early education content such as happy children's songs, Chinese classics, and enlightening English, and the cloud content can be continuously updated.