机器学习导论_机器学习导论

机器学习导论

什么是机器学习？ (What is Machine Learning?)

Machine learning can be vaguely defined as a computers ability to learn without being explicitly programmed, this, however, is an older definition of machine learning. A more modern definition was given by Tom Mitchell, "A computer program is said to learn from experience E with respect to some class of tasks T and performance measure P, if its performance at tasks in T, as measured by P, improves with experience E."

可以将机器学习模糊地定义为无需明确编程即可学习的计算机能力，但是，这是机器学习的较早定义。汤姆·米切尔(Tom Mitchell)给出了一个更现代的定义： “如果某计算机程序在T中的任务上的性能(由P来衡量)随着经验的提高而提高，则该计算机程序可以从经验E中学习一些任务T和性能指标P E。”

For instance, let's assume we have an algorithm that watches emails a user marks as spam and based on that observation it learns to filter out unwanted spam messages. The experience E in the above situation would be to Watch and recognize what type of mail is marked as spam. The task T would be to filter mail as spam based on the experience E. The Performance P would is the efficiency at which the algorithm filters spam mail and it would simply improve with the experience E.

例如，假设我们有一个算法可以监视用户标记为垃圾邮件的电子邮件，并根据该观察结果学会过滤掉不需要的垃圾邮件。在上述情况下的体验E是观察并识别哪种类型的邮件被标记为垃圾邮件。任务T是根据经验E将邮件过滤为垃圾邮件。性能P将是算法过滤垃圾邮件的效率，并且会随经验E的提高而提高。

Machine learning is often confused with Artificial intelligence. Artificial intelligence is measured as the ability of a machine to behave as a human being whereas Machine learning is a subset of artificial intelligence that deals with training a machine or computer to learn from large amounts of data supplied to it.

机器学习通常与人工智能相混淆。人工智能被衡量为机器表现为人类的能力，而机器学习是人工智能的子集，其处理训练机器或计算机以从提供给它的大量数据中学习。

Machine learning is implemented in two ways, Supervised and Unsupervised learning.

机器学习有两种实现方式，监督学习和无监督学习。

Supervised learning is when the machine is given a specific data set along with the correct output. Here the machine is given an idea of what the output must look like with respect to the given input. Supervised learning is further classified into two subsets namely, Regression learning problems and Classification learning problems.

监督学习是指为机器提供特定的数据集以及正确的输出。在这里，机器将获得关于给定输入的输出外观的概念。监督学习被进一步分为两个子集，即回归学习问题和分类学习问题。

In a regression learning problem, we try and obtain predictions as a continuous function of the given input and not as a discrete value whereas in Classification learning problems we try to obtain a discrete value of the output based on previously analyzed data and the given input.

在回归学习问题中，我们尝试获取作为给定输入的连续函数而不是离散值的预测，而在分类学习问题中，我们尝试基于先前分析的数据和给定输入来获取输出的离散值。

In classification learning problems, on the other hand, we approach problems without any knowledge about the correct output. The required relationship between the given data and solution can be acquired by clustering the given data based on the relationship of the individual variables present in the given data.

另一方面，在分类学习问题中，我们在没有任何正确输出知识的情况下处理问题。可以通过基于给定数据中存在的各个变量的关系对给定数据进行聚类来获取给定数据与解决方案之间的所需关系。

Machine learning is used and implemented in various fields of application. Most of us use machine learning algorithms unknowingly in our daily lives. Some of the common applications of machine learning are, Social media services such as personalized social media and news feeds by the content is being searched for, advertisement targetting and product recommendations by monitoring products or services viewed online, email and malware filtering by monitoring the content marked as spam and content classified as malware by users, Refining search engine results to improve search result by monitoring the time spent visiting and viewing web results, personalizing home and voice assistants by monitoring users internet and web activity. Machine learning is an important aspect to predicting highly accurate solutions to problems in various fields of applications such as science, medicine and commerce and can be employed to simplify and improve the quality and rate at which problems are solved.

机器学习在各种应用领域中得到使用和实现。我们大多数人在日常生活中不知不觉中使用了机器学习算法。机器学习的一些常见应用包括：社交媒体服务(例如按内容搜索个性化社交媒体和新闻提要)，通过监视在线观看的产品或服务来确定广告目标和产品推荐，通过监视内容来进行电子邮件和恶意软件过滤被用户标记为垃圾邮件和被用户归类为恶意软件的内容，通过监视访问和查看Web结果所花费的时间，通过监视用户的Internet和Web活动来个性化家庭和语音助手来完善搜索引擎结果以改善搜索结果。机器学习是预测诸如科学，医学和商业等各种应用领域中的问题的高精度解决方案的重要方面，并且可以用来简化和提高解决问题的质量和速度。