The core of artificial intelligence: 1. Computer vision refers to the ability of computers to recognize objects, scenes and activities from images; 2. Machine learning refers to the fact that computer systems do not need to follow explicit program instructions; 3. Natural language processing; 4. Robots; 5. Speech recognition, which mainly focuses on automatic and accurate transcription of human speech technology.
The operating environment of this article: Windows 7 system, Dell G3 computer.
The core of artificial intelligence:
1. Computer vision
Computer vision refers to computers identifying objects from images , scene and activity capabilities. Computer vision techniques use sequences of image processing operations and other techniques to break down image analysis tasks into manageable chunks. For example, some techniques can detect the edges and textures of objects from images, and classification techniques can be used to determine whether the identified features represent a class of objects known to the system.
Computer vision has a wide range of applications, including: medical imaging analysis is used to improve disease prediction, diagnosis and treatment; facial recognition is used by Facebook to automatically identify people in photos; it is used in the security and surveillance fields used to identify suspects; and when it comes to shopping, consumers can now take photos of products with their smartphones to gain more purchasing options.
Machine vision, as a related discipline, generally refers to vision applications in the field of industrial automation. In these applications, computers identify objects such as production parts in highly constrained factory environments, making the goals simpler than computer vision that seeks to operate in unconstrained environments. Computer vision is an ongoing research, while machine vision is a "solved problem", a system engineering topic rather than a research level topic. As applications continue to expand, some computer vision startups have attracted hundreds of millions of dollars in venture capital since 2011.
#2. Machine learning
Machine learning refers to the fact that the computer system does not need to follow explicit program instructions, but only relies on data to improve their own performance. At its core, machine learning automatically discovers patterns in data, and once discovered, patterns can be used to make predictions. For example, if a machine learning system is given a database of credit card transaction information such as transaction time, merchant, location, price and whether the transaction was legitimate, the system will learn patterns that can be used to predict credit card fraud. The more transaction data processed, the more accurate the predictions will be.
Machine learning has a wide range of applications, and it has the potential to improve the performance of almost any activity that generates huge amounts of data. In addition to fraud screening, these activities include sales forecasting, inventory management, oil and gas exploration, and public health. Machine learning technology also plays an important role in other cognitive technology fields, such as computer vision, which can improve its ability to recognize objects by continuously training and improving visual models in massive images.
Today, machine learning has become one of the hottest research areas in cognitive technology, attracting nearly US$1 billion in venture capital between 2011 and 2014. Google also spent $400 million in 2014 to acquire Deepmind, a company that researches machine learning technology.
#3. Natural language processing
Natural language processing refers to the human-like text processing capabilities that computers have. For example, extract meaning from text, and even independently interpret meaning from texts that are readable, natural in style, and grammatically correct. A natural language processing system does not understand the way humans process text, but it can skillfully process text in very complex and sophisticated ways. For example, automatically identify all the people and places mentioned in a document; identify the core topics of the document; extract various terms and conditions from a pile of human-readable contracts and create a table. These tasks are simply impossible to accomplish with traditional text processing software, which only performs simple text matching and patterns.
Natural language processing, like computer vision technology, integrates various technologies that help achieve the goal. Language models are built to predict the probability distribution of language expressions, for example, the maximum likelihood that a given string of characters or words expresses a specific semantic meaning. The selected features can be combined with certain elements in the text to identify a piece of text. By identifying these elements, certain types of text can be distinguished from other text, such as spam emails from normal emails. Classification methods driven by machine learning will become the filtering criteria used to determine whether an email is spam.
Поскольку контекст очень важен для понимания разницы между «таймингами» и «дрозофилами», практическое применение технологии обработки естественного языка относительно узко. Эти области включают анализ мнений клиентов о товаре, отзывы о конкретных продуктах и услугах, автоматическое раскрытие определенных последствий в гражданских исках или правительственных расследованиях, автоматическое написание шаблонных эссе о таких вещах, как корпоративные доходы и спорт, и многое другое.
# Интегрируйте когнитивные технологии, такие как машинное зрение и автоматическое планирование, в чрезвычайно маленькие, но высокопроизводительные датчики и исполнительные механизмы. Помимо продуманного оборудования, это привело к появлению нового поколения роботов, способных работать вместе с людьми и гибко решать различные задачи в различных неизвестных средах. Например, дроны, «коботы», которые могут делиться работой с людьми в цехе и т. д.
######5. Распознавание речи#########Распознавание речи в основном связано с автоматической и точной расшифровкой человеческой речи. Технология должна столкнуться с некоторыми проблемами, аналогичными обработке естественного языка, с некоторыми трудностями в обработке различных акцентов, фонового шума, различении омофонов/синонимов («купить» и «по» звучат одинаково), и в то же время вам также необходимо быть способен работать со скоростью, соответствующей нормальной скорости речи. Системы распознавания речи используют некоторые из тех же методов, что и системы обработки естественного языка, дополненные другими методами, такими как акустические модели, которые описывают звуки и вероятность их появления в определенных последовательностях и языках. Основные приложения распознавания речи включают медицинскую диктовку, голосовое письмо, голосовое управление компьютерными системами, обслуживание клиентов по телефону и т. д. Например, Domino's Pizza недавно запустила мобильное приложение, которое позволяет пользователям размещать заказы с помощью голоса. ######Для получения дополнительной информации посетите колонку ###FAQ###! ###The above is the detailed content of What is the core of artificial intelligence. For more information, please follow other related articles on the PHP Chinese website!