Recently, often come across the term "machine learning". My question consists of two parts. First, for what machine learning is used, if possible with the most interesting examples. And second, more exciting to me, what preliminary knowledge is needed in order to begin to study independently the “Machine Learning” course?

Closed due to the fact that it is necessary to reformulate the question so that it is possible to give an objectively correct answer to the participants of RiotBr3aker , andreymal , aleksandr barakin , 0xdb , AK ♦ 12 May at 19:20 .

The question gives rise to endless debates and discussions based not on knowledge, but on opinions. To get an answer, rephrase your question so that it can be given an unambiguously correct answer, or delete the question altogether. If the question can be reformulated according to the rules set out in the certificate , edit it .

    1 answer 1

    Wikipedia describes the practical applications of machine learning quite well:

    • Speech recognition
    • Gesture recognition
    • Handwriting recognition
    • Pattern recognition
    • Technical diagnostics
    • Medical Diagnostics
    • Time Series Prediction
    • Bioinformatics
    • Fraud Detection
    • Spam detection
    • Document categorization
    • Stock technical analysis
    • Financial supervision
    • Credit scoring
    • Predicting customer care
    • Chemoinformatics
    • Learning rankings in information retrieval

    What prior knowledge is needed in order to begin to study independently the "Machine Learning" course

    in my opinion, the following opinion is enough for a subjective opinion:

    • good motivation (this is, in my opinion, the main point)
    • understanding the basics of linear algebra. Good courses in machine learning (for example, a course from prof. Andrew Ng ) allow you to recall the necessary moments and smoothly lead you from the simplest to the most complex.
    • for practical application, it would not be bad to master basic coding skills in one of the programming languages. For Machine Learning most popular is Python . Java and R also common.
    • one
      MaxU-with all due respect, the separation of regression and classification tasks is not entirely correct. Both of these attempts are first to construct a model using known samples and then use it to predict the value of some previously unknown object. The difference in the type of data: for regression - the data are measured on a numerical scale, for classification - dichatomic or nominal data. In machine learning, all this is considered as a “learning with a teacher” task. And clustering is a completely different type of task - “learning without a teacher” - you need to take some objects and divide them into (unknown) classes. - passant
    • @passant, I agree - this is a controversial point. At first it was worth mentioning training with and without a teacher, and only then regression and classification. Removed this paragraph ...;) - MaxU
    • Thanks for the comments. Another question arose on motivation. What can it be for the student? Could you highlight in descending order, at least 5 points. Thank. - Emil Gariff
    • one
      @EmilGariff, this is for you here . By the way, to thank the author of the answer, you can vote for the answer (you must have at least 15 reputation points) and / or accept the answer as correct ;) - MaxU