What had once been an ordinary mobile application going by the name of Burbn, has now risen proudly to become worldwide fame.Yep…. I’m talking about Instagram.
Crossing a billion monthly active users recently, the photo and video sharing social network platform, Instagram has established its way towards becoming one of the most acclaimed and famous apps on a global scale. Started in 2010, the app has progressed swiftly, swamping its competitors, by consistently upgrading its features and engaging itself in constant evolution and has paved its way towards becoming the 5th most downloaded app as of the end of 2019.
Presently the app has created a high degree of frenzy among the youth of the day, particularly the escalating segment of people who prefer conveying their thoughts through images. With millions of posts being uploaded on the platform every day and users engaging with these posts regularly by liking, commenting and using hashtags, this is where big data enters the picture.
"Instagram isn't necessarily a photo company or a communications company as I like to say, we're also going to be a big data company” - As quoted by Kevin Systrom, CEO of Instagram
A huge volume of data is being generated through the constant unfolding activity and Instagram’s been making exemplary use of it, both in case of managing this data as well as in leveraging it with Artificial Intelligence algorithms to gain insights.
“The whole idea of machine learning is that it’s far better about understanding those nuances than any algorithm has in the past, or than any single human being could,” Instagram co-founder and then-Instagram CEO Kevin Systrom
Through the support of tags and trending information, the users can find photos and posts on particular topics or activities, events and also for exploring experiences, trending restaurants and places around the globe.
Basically Instagram recognizes accounts that are more or less similar to one another by adopting a machine learning technique termed as “word embedding”. This technique deciphers the order in which words appear in the text in order to measure how connected they are. Instagram uses the same technique to decipher and comprehend how connected any two accounts are to each other.
You can get a better understanding of machine learning by referring to our other blogs on the subject.
So in order to make its recommendations, the Explore system starts by observing the “seed accounts” which are the accounts the users have interacted with in the past by liking or saving their content. It then discovers the accounts that are similar to these and selects 500 pieces of content from them.
These content pieces are then filtered in order to remove all spam, misleading and policy-violating content from them and then the remaining posts are ranked on the basis of how probable a user is to interact with each one. At long last, the top 25 posts are then sent to the first page of the user’s Explore tab.
These bots are designed to automate the user’s account interactions. These do everything from liking comments that customers leave on posts, to posting comments on another person’s content. This is seen as an excellent way of increasing engagement. An example of an Instagram bot would be Kenji.AI.
This is a new Instagram bot which automates the user’s activity efficiently. It makes use of machine learning algorithms to comprehend who would be most likely to engage with your account.
Instagram ensures it uses the big data it generates fully to its advantage by extracting and analysing the customer insights it gains from it. Instagram sells advertising space to companies who happen to be interested in reaching a particular target audience and in sending out a particular marketing message by comprehending and understanding the search preferences and engagement insights of its users.
Being owned by a powerful tech giant like Facebook allows Instagram to have a vast network of insights and information for helping target advertising based on the audience’s likes, who they follow and engage with and the posts they save.
With the level of content shared on the app rapidly growing it becomes more and more paramount for the platform to deliver content that is relevant to its users. Hence in 2016, Instagram altered its feed to display first, the posts it believes it’s users would favour and share instead of in reverse- chronological order.
To do this machine learning algorithm was put to work to go through all the content and carefully comprehend which of the content would be more relevant for its users, in order to design a personalised feed for each of them.
With an abundance of content being shared every day across the app, some of it is bound to be spam.
Now how are these spam messages detected?
Instagram makes use of Artificial Intelligence’s text analytics algorithm “DeepText” for dealing with spam messages. Its spam filter can detect spam messages in over 9 languages that include English, Arabic, and Chinese. Once detected, these messages are automatically removed. The algorithm is able to comprehend a message’s context almost as well as humans.
“No one likes you!”
“You are ugly!”
“You’re a loser!”
Social Media has since long been the instrumental platform for people, particularly teenagers to indulge in cyberbullying and Instagram is no exception.
The platform has vowed to fight online bullying by leveraging artificial intelligence techniques that will foresee and recognise any kind of bullying or offensive texts on the platform. For the same, the platform has recently launched a new AI feature that works by keeping track of a list of words and phrases which have been reported offensive in the past and then alerting it’s users whenever their captions for a certain photo or video could be considered offensive in order to give them a chance to halt and re-assess their words before posting them.
Around the month of October 2019, Instagram launched a feature termed as “Restrict” that allows the platform’s users to easily shadowban any users that are posting offensive or bullying comments. This means that the comments on the posts of a person who has been shadowbanned by the user will only be detectable to that person.
Instagram with its millions of daily shared posts has the potential of becoming a beneficial cultural analysis tool.
For instance, As per a study launched by Cornell University in 2016, a group of researchers aimed at exposing how cultural clothing trends vary across the world, in particular, looking at fashion trends, based on their time and location, between 2013-2016.
Using techniques like face recognition to eliminate irrelevant photos, the researchers designed an object recognition program that could distinguish and recognize items of clothing, for instance, a shirt from a t-shirt or a jacket from a sweater. The program studied and determined what trends were popular in what areas, what clothes were being paired with what, and how the trends had varied within the time period of the research.
“Imagine a future anthropologist with access to trillions of photos of people - taken over centuries and across the world - and equipped with effective tools for analyzing these photos to derive insights,” - the team stated
“What kinds of new questions can be answered? This problem area of data-driven visual discovery is still new, but is beginning to gain attention in computer vision and graphics.”
Instagram has been making apparent use of its Big Data for crisis communication. For instance, during the 2012 Hurricane Sandy, social media, Instagram, in particular, was instrumental in connecting the victims of Sandy.
Thousands of images were embedded into tweets in the two week time periods. This helped perceive the change in affairs surrounding disasters while also helping the victims get through to their families as the people active on social media were able to see that their loved ones were safe, while also getting a good look inside the storm.
Yet another instance where Instagram played a crucial role was during the Ebola crisis. Various Instagram users, ranging from photojournalists to humanitarian organizations were able to show a unique side of the Ebola story. Amidst the tumult faced with the quarantine, the death of loved ones and the uncertainty of the future the images uncovered an array of sentiments from fear and frustration to beauty and hope.
With Instagram’s escalating popularity and acceptance, pops the question, what is making this app so popular? A lot of its growth and widespread acceptance has to do with its strategic use of data.
Yes, Instagram is indeed one of the renowned companies that put AI and Big Data to good use and its result speaks for the power these new-age technologies propagate. For more blogs on Analytics, Do read Analytics Steps!
Mallika is an eager and enthusiastic intern at Analytics Steps. Mallika believes that words hold the power to clarify and illuminate technicalities of various subjects and help readers in gaining understanding and knowledge. Her love for exploring and absorbing new technologies helps her keep pace with the ever changing digital world.
Introduction to Time Series Analysis: Time-Series Forecasting Machine learning Methods & ModelsREAD MORE
How is Artificial Intelligence (AI) Making TikTok Tick?READ MORE
The Essence of Game Theory in Artificial Intelligence - 5 Types of Game Theory and Nash EquilibriumREAD MORE
7 Types of Activation Functions in Neural NetworkREAD MORE
Convolutional Neural Network (CNN): Graphical Visualization with Code ExplanationREAD MORE
Deep Learning - Overview, Practical Examples, Popular AlgorithmsREAD MORE
6 Dynamic Challenges in Formulating the Imperative Recommendation SystemREAD MORE
Introduction to Machine Learning: Supervised and Unsupervised LearningREAD MORE
What are the roles, opportunities and challenges posed by Big Data in Tourism?READ MORE
Driving Digital Transformation with Data Science: What, How and Why?READ MORE