Uber

The mechanism tells where the demand of can services is more. The cabs are distributed in such locations. How does it decide the optimal position of cabs in a city. Uber will complete 12 years in 2021. Data, time, longitude and latitude is recorded each time. This data is feeder into model to tell the clustering groups. Machine learning is the subset of AI which gives the ability to perform task on their own. The machine can grow, learn and change predictions. Chatbots, speech recognition, image recognition is type of Machine Learning problem.

Supervised learning needs supervision. Every model has to be trained on some model so it can perform on that data. To train the model in such a way it is able to map the input to some output. We tell the machine the Question and the answer. We are training the machine learn on the past experience. The example of this is: Teacher guiding a student.

In unsupervised learning we don’t tell the computer anything. We just pass the data to the model and the model based on its own experience comes with an output. Based on it the model predicts or classifies. Uber uses unsupervised learning algorithm. The loophole is the machines don’t have scope of learning. It works in raw data. It can find trends and patterns without human intervention. U

Uses clustering to group location, cab services, etc.

Clustering algorithm: Helps divide data points into small clusters. It’s used in fake news detection, classification of research papers into categories and spam detection. I

K- means clustering: K refers to number of clusters we want to draw out from our data. Find the mean.

Centroid initialisation-> clustering -> updating centroid