{"id":3600,"date":"2018-06-22T11:47:59","date_gmt":"2018-06-22T10:47:59","guid":{"rendered":"http:\/\/insideintercom.io\/?p=3600"},"modified":"2020-07-30T12:55:37","modified_gmt":"2020-07-30T11:55:37","slug":"machine-learning-way-easier-than-it-looks","status":"publish","type":"post","link":"https:\/\/www.intercom.com\/blog\/machine-learning-way-easier-than-it-looks\/","title":{"rendered":"Is machine learning hard? Not always"},"content":{"rendered":"

There\u2019s a common misconception that you have to be an AI wizard or mathematician to use machine learning in your work. That machine learning requires hard calculus.<\/p>\n

After all, you\u2019re teaching machines that work in ones and zeros to reach their own conclusions about the world. You\u2019re teaching them how to think!<\/p>\n

But like many frameworks we have for understanding our world \u2013 Newton\u2019s Laws of Motion, Jobs to be Done<\/a>, Supply & Demand \u2013 the fundamental ideas at the heart of machine learning are relatively straightforward.<\/p>\n

Our own machine learning expert, Fergal Reid<\/a>, describes it this way: “You have a problem, you\u2019re trying to solve it, and then you have a system where the performance improves when you give it more training data…The more data you get, the better your estimate.”<\/p>\n

If you\u2019ve got a well-defined problem that can be solved given enough time and example data, you should give machine learning a try. Here\u2019s an example.<\/p>\n

Example problem \u2013 without using machine learning<\/h2>\n

Say we want to include a \u201cYou might also like\u201d section at the bottom of this post. How should we go about doing this?<\/p>\n

\"Machine<\/div>\n

If you\u2019re avoiding machine learning at all cost, you might opt for this approach:<\/p>\n

    \n
  1. Split the current post title into its individual words.<\/li>\n
  2. Get all other posts.<\/li>\n
  3. Sort all other posts by those with the most words in their body in common with our title.<\/li>\n<\/ol>\n

    Or, in Ruby:<\/p>\n