Machine learning is a powerful machine intelligence technique that can be used in a variety of settings to generate data insights. In this blog post, we will explore real-world or real-life machine learning / deep learning / AI examples from daily life. We’ll see how machine-learning techniques have been successfully applied to solve real-life problems. The idea is to make you aware of how machine learning and data science applications are everywhere.
What are some real-world examples of machine learning from daily life?
Here are some real-world examples of machine learning that we use in our daily life:
- Best driving directions (Google Maps): A bunch of machine learning / deep learning models are used to determine the best (shortest) route between places across different cities in different countries.
- Uber, Ola taxi rates: Machine learning algorithms are used to continuously determine the best (affordable) price for a taxi ride ensuring win-win for both taxi owner / driver and the customers. Machine learning algorithms such as multi-armed bandit algorithms are used to achieve this.
- Recommendation in news-feed (Facebook): Have you wondered whether you are shown the post in your Facebook account based on what you have been seeing in past? Facebook uses machine learning algorithms such as collaborative filtering and PageRank algorithm to recommend content on your news feed based on the interest you show in them or similar users like you. For example, if one of your friends liked a particular post by another friend in Facebook, machine learning algorithms recommend you also like that similar post.
- Netflix or Amazon Prime movie recommendations: Machine learning models help suggest you new movies based on your history and ratings of previous films. In addition, it does recommend movies based on similar user profiles. Machine learning algorithms such as collaborative filtering are used to build this model. There are various apps such as Netflix, Amazon prime, Disney Hotstar etc which are using machine learning / deep learning for recommending the movies.
- Order arrival time/ETA (Food ordering): Machine learning models are also used to predict the time of arrival (ETA) for your food order. For example, you can see estimated wait times on Swiggy, Zomato, UberEATS app etc. Algorithms such as Hidden Markov Model (HMM) and Dynamic Time Warping (DTW) are used to make real-time ETA for food orders.
- Order arrival time / ETA (eCommerce): eCommerce machine learning models predict when your order will arrive. For example, Flipkart, Amazon, Walmart or similar eCommerce product pages provide estimated delivery dates and timeframes for different products. Machine learning algorithms such as HMM and DTW are used to make real-time ETA for eCommerce orders.
- Discount (eCommerce): Machine learning models can also predict customers’ buying behavior and provide them with appropriate discounts. The machine learning model essentially helps in identifying the probability of when / if will you buy a product or not, which is based on various factors such as recency (past purchasing history), frequency (how many times have you purchased from this store) etc. Algorithms such as Bayesian Networks, Linear Regression are used to check the likelihood of a customer buying or not.
- Product recommendation (eCommerce): Machine learning models recommend products to customers for their next purchase based on the customer’s previous history and other similar users’ behavior. This is one popular solution across almost all eCommerce websites. Machine learning algorithms such as collaborative Filtering are used to build this machine learning model.
- Alexa, Siri, Google voice assistant: Machine learning models are used to power voice assistants like Alexa, Siri and Google Assistant. Using machine learning algorithms, the machines can understand spoken language queries very well. Machine learning algorithms such as machine translation, natural language processing are used to build machine learning voice assistants.
- Chatbots on different websites / apps: You might have come across chatbots available on ecommerce, food ordering, and many other apps/websites. Almost all of them are AI-powered chatbots. Chatbots powered by machine learning / deep learning help answer questions in real-time across different channels (messengers) such as Facebook Messenger, Skype etc.). Machine learning algorithms such as machine translation, natural language processing are used to build machine learning chatbots.
- Automated/Robo phone calls: You may have come across automatically dialed (robocalls) and machine-generated phone calls. Machine learning models are used to understand your real interest in a particular call, which is based on various factors such as tone of voice / language etc., so that you get the most relevant telemarketing calls for your business. Machine learning algorithms such as machine translation, text-to-speech, natural language processing models are used to build machine learning telemarketing models.
- Spam emails detection: Have you observed that email services such as Gmail, Yahoo etc has been very successful in filtering out spam emails from your email box. It is machine learning classification models which is running underneath. Machine learning algorithms are used to help detect and separate the important emails from the unimportant ones (spam).
- Self-driving cars/trucks: Self-driving vehicles powered by machine learning / deep learning is one of the most successful and real machine learning examples in the real world. Self-driving cars and trucks help reduce human effort, accidents and other related problems. A bunch of deep learning and machine learning models are used to build self-driving cars.
- Detecting fraud transactions: Machine learning models can be applied to detect fraudulent transaction or machine / device behavior in financial services, telecom industry etc. For example, machine learning algorithms such as Random Forest (RF), Support Vector Machines (SVM) & K-Nearest Neighbors (KNN) are used to detect fraud transactions.
Simple introduction to Machine Learning
Simply speaking, machine learning represents the phenomenon of enabling machine to learn mathematical function (approximate) which can be used to predict to deal with real-world scenarios. Do you remember function like the following. This is just one simple equation:
Y = 3A + 5B – 9
In the above equation, “3” and “5” and “9” have always been given to us when we used to solve these kind of equations in our school days. In case of machine learning, it is these “3”, “5” and “9” that is learned with the help of historical data set. In addition, the quation such as above can be one of the equations. Other equations such as the following can also be learned:
Y = 3SinA + 5B + 9C – 10
Y = 6A*B – C + 3
Here is a definition of machine learning from my other post – Machine learning concepts and examples:
Mathematically speaking, machine learning is about approximating mathematical functions (equations) representing real-world scenarios. These mathematical functions are also referred to as “mathematical models” or just models. Thus, machine learning models are mathematical equations/functions that represent real world problems/scenarios. The reason why machine learning models are called function approximations because it will be extremely difficult to find exact function which can be used to predict or estimate real world scenarios.
Machine learning is all around us. It’s not just something that machine learning professionals use to solve the world’s most complex problems, it’s also a technology so integral to our lives that we take it for granted. Whether you’re using Siri or Google Assistant, ordering your groceries online with chatbots, getting automated/robocalls from telemarketers who are able to identify which customers will be interested in their products based on machine-learning algorithms that analyze data like tone of voice and language patterns – machine learning has infiltrated into every aspect of life today. If you want to learn more about machine learning, please drop a message.
- Positively Skewed Probability Distributions: Examples - March 21, 2023
- Maximum Likelihood Estimation: Concepts, Examples - March 20, 2023
- Generative Modeling in Machine Learning: Examples - March 19, 2023
Leave a Reply