It is not an exaggeration to say that the world we live in is getting redefine by Data Science. For Example:
- We use Amazon/any e-commerce sites to shop
- thus, We use Uber/any travel aggregators to travel
- We use Siri/Alexa/Google Assistant on our smartphones to ask questions and do tasks
- so, We use Facebook/LinkedIn/any social network for connecting At the backend of every one of these platforms is Data Science.
However, even if you haven’t used any of these, you are a beneficiary of Data Science you:
- Have taken a flight
- Are driving a car
- Are using any FMCG product
Let’s look at some general use cases of Data Science in various organizations:
Amazon
- Alexa is build on far-field speech recognition. Far-field speech recognition involves interpreting voice commands spoken from a microphone at a distance. Such commands are usually distorted due to ambient noise, and it is a challenge to recover the intended command. Moreover, the recognition and the corresponding action must have a low turnaround time. Next time you ever get to say ‘Hey! Alexa’, think of the Data Science going on in the backend.
- Amazon Go – Walk in and walk out of a store without stopping near the cashier. The cashierless checkout store at Amazon headquarters has a lot of Data Science going on in the background. It uses different techniques to figure out which items a customer is taking in and out of a shelf. Once the customer leaves the store, the final bill is calculate by and charge up against the customer’s Amazon account.
- Google Assistant is now multilingual. Irrespective of the language instructions are giving in, the Assistant will figure out what you are trying to say. Jump between two languages seamlessly across queries without going back to the language settings. Identifying the language being spoken and then figuring out the query is quite difficult, and it involves a lot of Data Science in the backend.
- Smart Linkify: In this new Android Pie feature, when you type an address in the text, it automatically detects the address and provides a link to the address in Google Maps. The main challenge is detecting the entity of the address within the text as well as identifying when the address begins and when it ends in the entered text.
A general Facebook feed contains a lot of images and the ability to discuss and comment on the visual imagery forms an integral part of a user’s experience. But this experience is not accessible to visually impaired users. To make them understand better, augmenting the visual content in their feed, Automatic Alt-text was conceived.
As part of this project, a small audio snippet describing the content of images is automatically generate.
Being a social network with rich interconnections and interactions, Facebook provides an ideal field to study social interactions and ask interesting sociological questions and find their answers.
Uber
An interesting use case of Data Science at Uber is the restaurant delivery estimate for UberEats. The Data Science models predict how much time it will take to prepare a meal and it’s delivery time before the order is issue, and predicting each step of the delivery process.
Food delivery is a multi-step process right from a restaurant acknowledging an order to food preparation to the assignment of a delivery person to delivery at doorstep. To make a decent prediction end-to-end and readjust these predictions at every step of the process, lots of Data Science processes happen in the backend.
Airbnb
Airbnb is a rental marketplace which contains pictures of houses and rooms that are available for the user to rent. But many times, these pictures are not tagged properly to inform the user about which part of the house it belongs to. The data science team classified the different parts of the home as – bedroom, living room, etc to inform the users about listings in a better manner and increase the chances of a house being picked up for a stay.
however, Another interesting use case is to predict the optimal rental pricing of a house. For tourist destinations, the closer you book the home to a special occasion like New Year’s eve, the higher the price will be. Even the hosts can be given information about the right pricing strategy – with insights into trends pushing the rent higher or lower.
IBM Watson
IBM Watson has entered into a partnership with US Open to enrich user experience during the US Open. An interactive chatbot has been deployed along with the US Open app. This bot can understand natural language questions and answer them in real-time. IBM Watson analyzes the match footage and automatically generates the highlight moments of play.
These use cases are just a small sample of a large world. These are the general use cases for tech companies. What about other industries? Data Science is ubiquitous in the true sense of the word and is making its way to core industries too.
Impact of data Science in Other Industries
We have seen the impact of Data Science in the technology landscape, now let’s look at its impact on various other industries ranging from aircraft to pharmaceuticals and financial markets.
Airbus
Airbus is using data science to process data coming from various factories and predicting when there is a change in the manufacturing process so that possible problems could be preemptively detected and solved. They also use Data Science in other applications such as flight operations optimization and virtual assistants for flight crew, and to also leverage social media to understand passenger experience.
Goldman Sachs
Goldman Sachs uses data science to get insights into investment themes such as momentum, value, and profitability. For insights on momentum, Data Science techniques are applied to identify connections between the companies based on industry sentiment and stock movements. For predictions on value, analysts mine industry-specific data that exists beyond the company’s financial statements.
Here, we have seen how, in other industries too, Data Science is playing a major role to improve processes. And many many startups are using Data Science under their hood to improve their product offerings.
written by: Thomas Vengazhiyil Alex
reviewed by: Kothakota Viswanadh
If you are Interested In Machine Learning You Can Check Machine Learning Internship Program
Also Check Other Technical And Non Technical Internship Programs
