We have all seen movies depicting different forms of artificially intelligent creatures, ranging from robots to aliens and sometimes just a moving hand which helps Tony stark by doing “odd” jobs.
As many articles previously on this blog have already given a gist about artificial intelligence, I will try to take this article a little further into the domain of machine learning.
When we were little kids, many of us were taught the alphabets in kindergarten and then using those alphabets we formed words and then using those words we formed sentences.
Machine learning works in a similar manner. In this system, the machine learns from a previously defined set of information which is known as the training set. Without further ado lets move onto defining the two types of machine learning methodologies.
Supervised Learning
Algorithms are trained using labelled examples, such as an input where the desired output is known. For example, a piece of equipment could have data points labelled either “F” (failed) or “R” (runs). The learning algorithm receives a set of inputs along with the corresponding correct outputs, and the algorithm learns by comparing its actual output with the correct outputs to find errors. It then modifies the model accordingly
Through methods like classification, regression, prediction and gradient boosting, supervised learning uses patterns to predict the values of the label on additional unlabeled data.
Supervised learning is commonly used in applications where historical data predict likely future events. For example, it can anticipate when credit card transactions are likely to be fraudulent or which insurance customer is likely to file a claim.
Let me try to explain the above using an example, Recently I started learning German and it was left to my teacher to take out all the French and Italian and fit in German into my brain. Believe me, it is a difficult task.
So, what my teacher did was, he decided to not start by explaining in English and then converting it to German, instead he started off directly with German, using sentences which has similar tones and sounds to that of English, which my brain easily picked up and created a data sync with.
Supervised Machine learning works French, we give the computer a large set of data which is already in the required format and help it learn what exactly we are looking for. Once the machine realizes what we are looking for, it hunts for this data in a large collection of random data which can be of any format.
Unsupervised learning
This is used against data that has no historical labels. The system is not told the “right answer.” The algorithm must figure out what is being shown. The goal is to explore the data and find some structure within.
Unsupervised learning works well on transactional data. For example, it can identify segments of customers with similar attributes who can then be treated similarly in marketing campaigns.
Or it can find the main attributes that separate customer segments from each other. Popular techniques include self-organizing maps, nearest-neighbour mapping, k-means clustering and singular value decomposition. These algorithms are also used to segment text topics, recommend items and identify data outliers.
The above might have seemed like a long haul for a short task, let me try to explain this in a simpler manner.
Let me take you’ll back to the first time I read an Italian novel, This was when I did not know Italian. I went through the first paragraph, it made absolutely no sense, then I went through the second paragraph and it made even less sense.
This was most probably since I did not know the language, but who knows. So I went on reading till the end of the page at which point I realised that I cannot understand Italian (Slow learner).
The next thing I did was to watch an Italian movie with subtitles (In Italian). I went through the whole of the movie, word by word, gesture by gesture and tried to map each gesture and emotion to the word which was being printed in the subtitles (Don’t judge me, I was jobless).
By the end of the movie, I could easily distinguish between words that expressed feelings such as that of joy and sadness etc. Now, I decided to go through the movie with English subtitles and to my surprise, 80% of my guesses were right and I had learnt how to labe expressions.
Unsupervised learning is something similar, the machine does not have any labels and works purely on the basis of the information provided to try to map on its own what it could mean. There are many algorithms utilized for this as mentioned before.
This article just introduces one to the surface of machine learning and I shall be writing more articles taking you’ll further into the concept of machine learning, only till where I know. As the name suggests, machine learning is just a machine trying to learn the human ways.
Further Reading: