10 Examples of machine learning

What is machine learning?

What is machine learning? The answer to this question can refer to the authoritative machine learning definition, but in reality machine learning is defined by the problem it solves. Therefore, the best way to understand machine learning is to observe some examples.

Take a look at some examples of machine learning problems that are well known and understood in real life, and then discuss the classification of standard machine learning problems (naming systems), and learn how to identify which standard case a problem belongs to. The point of doing this is to understand the type of problem we are facing and we can think about the data we need and the algorithms we can try.

　　10 examples of machine learning problems

Machine learning problems are everywhere, and they make up the core or hard part of the network or desktop software that is used daily. Twitter on the "want to try" advice and Apple's Siri voice understanding system is an example.

　　Here are ten real examples of what machine learning really is:

• spam detection : Identify what is spam and what is not, based on the messages in your mailbox. Such a model can help categorize spam and non-spam messages by programs. This example, we should not be unfamiliar.
• Credit card fraud detection : According to the user within one months of credit card transactions, identify which transactions are the user operation, which is not. Such a decision model can help the program to return those fraudulent transactions.
• Digital recognition : According to the handwritten zip code on the envelope, the number represented by each hand-written character is recognized. Such a model can help the program to read and understand the handwritten zip code, and classify letters according to geographical location.
• speech recognition : From a user's discourse, determine the specific requirements of the user. Such a model can help the program to be able and try to automatically populate the user's needs. IPhone with the Siri system has this feature.
• Face recognition : Identify photos that contain a person based on the many digital photos in the album. Such a decision model can help the program manage photos based on the face of the person. Some cameras or software, such as IPhoto, have this feature.
• Product recommendation : Identify which of these are the products that the user is genuinely interested in and willing to buy, based on a user's shopping history and lengthy list of favorites. Such a decision model can help the program to provide advice to customers and encourage product consumption. Sign in to Facebook or Googleplus and they'll recommend users who might be associated with it.
• Medical Analysis : predict what disease the patient may be suffering from, based on the patient's symptoms and an anonymous patient data database. Such a decision-making model can provide support for professional medical professionals.
• Stock Trading : Determine whether the stock is open, open or Jiancang according to the current and previous price fluctuations of a stock. Such a decision model can help the program provide support for financial analysis.
• Customer Segmentation : Based on the user's behavior patterns during the probation period and the behavior of all users in the past, identify which users will be converted into the product of the payment users, which will not. Such a decision model can help the program with user intervention to persuade users to make early payments or better participate in product trials.
• shape Identification : Judging the shape the user wants to portray, based on the user's hand-drawn on the touch screen and a known shape database. Such a decision model can help the program display the ideal version of the shape to draw a clear image. IPhone app Instaviz can do that.

These 10 examples show a good idea of what a machine learning problem is. There is a special anthology documenting the historical examples. An example of this is a decision that needs to be modeled, and an effective automated modeling of that decision brings benefits to an industry or area.

Some of the problems are in artificial intelligence, such as natural language processing and machine vision (dealing with issues that people can easily handle), the most difficult problems. Others are also difficult, but they are also classic machine learning issues such as spam detection and credit card fraud detection.

Think about the interactions you've had with online or offline software over the past week. You can easily speculate on 10 or 20 examples of machine learning that are used directly or indirectly.

　　Types of machine learning problems

With the examples of machine learning problems mentioned above, you must have been aware of some similarities. This skill is valuable because it is good at seeing the nature of the phenomenon, enabling you to think efficiently about the data you need and the types of algorithms you can try.

There are some common classifications for machine learning. The following classifications are typical of most of the problems we encounter when we study machine learning.

• classify : Tag data, that is, classify it into a category, such as spam/non-spam (mail) or fraudulent/non-fraudulent (credit card transactions). Decision modeling is to mark new unlabeled data items. This can be seen as identifying problems, modeling differences or similarities between groups.
• regression : Data is marked with a real value (such as a floating-point number) instead of a label. Easy-to-understand examples such as timing data, such as stock prices that fluctuate over time. The decision to model this is to estimate the value of the new unpredictable data.
• Clustering : Data is not tagged, but data can be grouped according to similarity and other measurements of the natural structure in the data. One example can be cited from the list of 10 examples: Managing photos based on faces rather than names. This way, users have to name groups, such as IPhoto on a Mac.
• Rule Extraction : The data is used as the basis for extracting the proposed rule (premise/result, aka IF). These rules, probably but not all, are pointing, meaning that these methods can identify statistically compelling relationships between the attributes of the data, but not all of which involve the need for prediction. There is an example of the relationship between buying a beer or buying a diaper, (this is the private rules of data mining, whether it is true or not, and both expectations and opportunities are articulated).

When you think of a problem as a machine learning problem (such as a decision problem that needs to be modeled from data), think about what type of problem can be borrowed directly, or what the user or demand expects, and vice versa.

　　Resources

There are few resources to list the problems of machine learning in the real world. Or maybe they were there, but I didn't find it. I still find some cool resources for your reference:

• annual "Humies" award : This is a number of awards that are awarded to those who calculate the results comparable to human algorithms. These algorithms are only working on data or paid functions, so they can be creative enough to violate patents. It's amazing!
• ai effect : There is the idea that as long as the AI program has achieved good results, it is no longer considered artificial intelligence, but only as a technology, and then used daily. This concept is also applicable to machine learning.
• AI Contest : This competition involves a very difficult problem in the field of artificial intelligence, and if these problems can be solved, it will be a powerful case of proving artificial intelligence (the kind of real AI that is imagined in science fiction). Both computer vision and natural language processing are examples of AI race problems, and they are also considered to be the domain-specific classification of machine learning problems.
• 2013 Machine Learning Ten Questions : This Quora question has some very wonderful answers, one of which lists a rough breakdown of the actual machine learning problems.

We discussed some common examples of machine learning problems in the real world and their types. Now, we have information about whether a problem is a machine learning problem, and we can pick out some elements from the problem description to determine if it belongs to a classification type, a regression ray system, or a rule extraction type.

Do you know some machine learning problems in the real world? Comment and share your thoughts.

Bole Online-Victoria

Link: http://blog.jobbole.com/62334/

10 Examples of machine learning