Statistics for AI: Mean, Median, Mode, Variance and Standard Deviation

Hello students, welcome to Kwickprep. Two classes have the same average marks, but one class has toppers and failures, and the other has everyone in the middle. How can AI tell them apart? Today we learn why AI needs maths, then mean, median, mode, variance and standard deviation, and where statistics is used in real life.

Let us begin with why AI needs maths at all. An AI system turns photos, words and sounds into numbers, and learns patterns from them. Statistics helps summarise huge amounts of data into a few useful numbers. Probability helps AI predict what is likely to happen next. And maths lets us measure how wrong a model is, so it can improve.

First, three words we will use again and again. Statistics is the branch of maths that collects, organises and analyses data to find meaning. Central tendency means a single value that represents the middle, or the typical value, of the data. Dispersion means how spread out the values are from that middle.

We will use one example for the whole lesson. Asha, Ben and Chetan scored two, four and four out of ten. Divya and Esha scored four and five. Farhan, Gita and Hari scored five, seven and nine. So there are eight students, and the marks add up to forty.

The mean is the average, and it is the most common measure of central tendency. Step one, add all the marks, which gives forty. Step two, count the values, and there are eight students. Step three, divide the sum by the count, forty divided by eight, so the mean is five.

The median is the middle value when the data is sorted. First, arrange the marks from smallest to largest. There are eight values, an even count, so there are two middle values, the fourth and fifth, which are four and five. The median is the average of these two, so four plus five, divided by two, gives four point five.

The mode is the value that appears most often. Count each mark. Four appears three times, the most. Five appears twice. Two, seven and nine appear only once each. So the mode is four.

Pause and predict. Suppose Hari's score was forty nine instead of nine, because of a typing mistake. Which changes more, the mean or the median? The sum becomes eighty, so the mean jumps to ten, but the median stays four point five. This shows that the mean is pulled by extreme values, called outliers, while the median is not.

Now back to our opening question. Class A has all eight students scoring exactly five, so the mean is five. Class B is our class, with marks from two to nine, and its mean is also five. The means are equal, but the spread is very different, so we need a measure of spread.

Variance measures how far values are from the mean, on average. For each mark, subtract the mean, five, and square the answer, so negative gaps become positive. Two minus five is minus three, and its square is nine. Each four gives minus one, and a square of one. The two fives give zero. Seven and nine give two and four, with squares four and sixteen. All the squares add up to thirty two.

Now divide the total by the number of students to get the variance. Thirty two divided by eight gives a variance of four. Standard deviation is the square root of the variance, so the square root of four is two. Class A, where every mark is five, has a standard deviation of zero, because there is no spread at all. Some textbooks divide by the count minus one for a sample, but here we used the whole class.

How do we read standard deviation? A small value means most values are close to the mean. A large value means the values are widely spread. It is in the same unit as the data, here marks, so it is easy to explain. For our class, a typical student scored about two marks away from the mean of five.

Statistics is used everywhere around us. In sports, batting averages, strike rates and consistency help selectors and teams choose players. In disaster management, rainfall and wind data help weather departments warn people before floods and cyclones. In health, the number of cases and symptoms helps doctors and governments predict how a disease may spread. AI tools use these same statistics to make faster predictions.

Here is a quick cricket example. Batter A scores fifty, fifty and fifty, so the mean is fifty and the standard deviation is zero. Batter B scores zero, fifty and one hundred, so the mean is also fifty, but the standard deviation is about forty one. Both average fifty, but batter A is far more consistent, and statistics shows it clearly.

Let us revise what we learned today. AI learns patterns from numbers, so it needs maths. The mean is the sum divided by the count, the median is the middle value, and the mode is the most frequent value. In our class, the mean was five, the median four point five, and the mode four. Variance and standard deviation measure spread, and our standard deviation was two. Statistics helps in sports, disaster warnings and disease prediction. Try finding these for your own test marks.