The Data Analytics Blog

Our news and views relating to Data Analytics, Big Data, Machine Learning, and the world of Credit.

All Posts

The Apps On Your Mobile That Use Machine Learning Algorithms

August 4, 2017 at 3:25 PM

 Seems like the term Machine Learning is popping up in mainstream media as the next big thing. The fact is, however, that Machine Learning went mainstream a long time ago. You don’t think so? Check your mobile phone. Chances are you’ve been using and benefiting from Machine Learning algorithms all this time without even knowing it. 

Read our opinion piece: Machine Learning Is Here To Stay

In this blog post, I go through some of the many apps on your mobile phone that use Machine Learning algorithms to make recommendations, get you to your destination quickly and safely, improve your photos, tell you what song you’re listening to and more. You’ll see, Machine Learning is not so far away. It’s already in the palm of your hand.

If you want to know what machine learning is, read our blog post on What is Machine Learning? And Other FAQs We Get...

Soundhound and Shazam

Music recognition apps like Shazam and Soundhound use machine learning to recognise the song you’re listening to. By analysing millions of songs and extracting features that are characteristic of each song and then storing it in a database, it is then able to compare the characteristics of the song you’re are listening to and compare it to the songs it has in its database until it finds a match. The more versions of the song it listens to (acoustic, instrumental, etc.) the better it becomes at identifying - to the point where it can even identify a song by someone simply humming it.

Spotify and Pandora

Music streaming sites like Pandora and Spotify use machine learning in a similar way that Amazon use it for cross-selling. By analysing the listening behaviour of millions of users and finding people with similar listening behaviour, machine learning is able to then able to make recommendations and compile personalised mixes based on the preferences, behaviour and compiled mixes of similar users.

Read our blog post here on how Netflix use Machine Learning to predict your next binge watch.

Google Translate

Google Translate uses machine learning to translate over 100 languages into natural text with good syntax and grammar. How it learns to do this is by analysing millions of translated documents from the web to learn vocabulary and identify patterns in the language. When it comes time to translate, it takes what it has learned and picks the most probable translation.

What’s even more impressive is that through its recent introduction of Google Neural Machine Translation, the neural network system now powering Google Translate, it can learn to translate between two languages it has never translated by taking what it has learned already from translating between other languages and either of those two languages.

So, if it already knows that “Si” in Spanish means “Yes” in English and that “Yes” in English means “Oui” in French, then it will logically determine that “Si” in Spanish means “Oui” in French despite it never having translated between Spanish and French before. 

Tinder

Tinder, the online dating app, uses machine learning to help you get more swipes to the right (agree to a connection) versus swipes to the left (decline a connection) by the choosing your best profile photo to display. Through A/B testing - swapping photos displayed to  other users to see which one elicits more swipes to the right - Tinder's algorithm not only learns which photo will get you more connections, but also in general which photo attributes tend to be perceived as negative, e.g. wearing hats or sunglasses and covering part of your face are a no-go, photos with friends or group photos and not smiling in a photo all tend to get fewer swipes right.

Also, the algorithm chooses which of your photos to display based on other users' specific behaviour. For example, if a user tends to swipe left on photos showing other users with cats or dogs, it will swap your photo with Fluffy or Fido - even if it's your preferred profile photo - with another one  in your collection. 

Siri, Cortana, and Google digital assistants

Apple, Google, and Microsoft’s digital assistants use machine learning for voice recognition and to respond intelligently to your requests. By listening to the requests being spoken into mobile phones by their millions of users worldwide and responding with what they think are the best answers or responses they learn through positive or negative reinforcement each time based on users’ actions, e.g. repeating or re-phrasing of request or not. So, the more you engage with your digital assistant the more it learns and gets to know you and becomes better at responding to your particular requests. 

Pinterest, Instagram, Facebook and LinkedIn

All of these social networks use Machine Learning to increase the relevance of feeds by prioritising the most relevant posts on a user’s home feed. By tracking “likes” and “shares” from users, the machine learning algorithms continually learn what posts users like and then prioritises similar posts - listing them first on a user's feed - to increase user engagement.

Machine learning is also being used to detect spam users and spam or offensive content, predict ad performance and relevance, to determine which emails to send to users, and to predict and prevent churn.

Google Photos

Google Photos’ Suggested Sharing feature uses machine learning to suggest sharing certain photos with certain people based on who is in the photo, where the photo or video was taken, or who was with you at the time. Of course, this also means that machine learning is being used for facial recognition to detect the users in the photo whom it thinks you should share the photo with. Radical!

But that’s not all. Google Photos’ also automatically creates albums for you from a collection of photos taken during a specific period and uses machine learning to automatically populate it with only the “best” photos selected by its algorithm as being in focus and well-composed. And one more thing, Google Photos’ machine learning can also recognise over 255,000 landmarks in your photos – from the real Eiffel Tower in Paris to the mock Eiffel Tower in Vegas.

Snapchat

Snapchat - the  mobile app that allows users to capture and share temporary videos and photos that disappear after a few seconds -   uses machine learning algorithms to detect people’s faces and apply a fun filter on a face.

The algorithm has learned how human faces are constructed – with the eyes, nose, mouth and other features all usually being in the same place in relation to each other. It can then superimpose the filter onto your face pretty accurately.

Snapchat are also using machine learning for ad targeting, or Goal based Bidding, by predicting which users are most likely going to respond to an ad. 

Uber and UberEats

Uber uses machine learning algorithms to predict the ETA (estimated time of arrival) of your ride. By analysing information from millions of trips, it is able to estimate the time it typically takes for a car to reach your destination from its point of departure at any given point in the day. Plus, by analysing data from millions of pick-up experiences, Uber’s machine learning algorithms will suggest walking to a “low friction” pick-up point like a nearby corner where there are no problems stopping to improve the pick-up experience for both rider and driver.

Uber is also using a similar algorithm for UberEats – its food delivery service - but taking it one step further to improve the accuracy of its predicted food delivery time. By analysing historic data of previous orders taken and adding on the typical food preparation time required once an order is taken, UberEats’ delivery time estimates are getting more accurate every day and allowing it to offer cool new features like displaying 30 minute delivery options and popular options near you based on past orders and details like time of day and delivery location. 

Gmail

Google says, that about 50 – 70% of the messages Gmail receives are spam. Thank goodness it’s been using machine learning for years to analyse spam and detect patterns that it can now use to identify and eliminate 99.9% of that spam before it reaches your inbox. 

Google Maps and Waze

Google Maps and Waze (owned by Google) navigation apps are running on millions of mobile devices at any given time sending users’ location and driving speed to a central server to predict driving times as well as alternate routes. Waze encourages users to drive with the app open even if they are not using it to calculate average speeds, check for errors, improve layout and learn road and turn direction. Waze collects data for every route driven with the app open, learns these routes and is able to suggest optimal routes based on what it’s learned from observing all users driving similar routes.

Whether you realise it or not, you are already using or benefiting from Machine Learning algorithms every day in various real-world and digital world ways. But I hope the above mobile app use cases bring machine learning now closer to home for you – to the palm of your hand - and get you thinking of the many cool ways you could apply machine learning to improve the profitability of your business and experience for your customers.

If you’re interested in bringing in some of that machine learning magic to your business, please get in touch to hear what our team of data scientists could do for you or learn about our Machine Learning as a Service for businesses, Genius.

Using machine learning in business - download guide

Julian Diaz
Julian Diaz
Julian Diaz was Head of Marketing for Principa until 2017, after which he became Head of Marketing for Honeybee CRM. American born and raised, Julian has worked in the IT industry for over 20 years. Having begun his career at a major software company in Germany, Julian made the move to South Africa in 1998 when he joined Dimension Data and later MWEB (leading South African ISP). Since then, Julian has helped launch various South African technology brands into international markets, including Principa.

Latest Posts

The 7 types of credit risk in SME lending

  It is common knowledge in the industry that the credit risk assessment of a consumer applying for credit is far less complex than that of a business that is applying for credit. Why is this the case? Simply put, consumers are usually very similar in their requirements and risks (homogenous) whilst businesses have far more varying risk elements (heterogenous). In this blog we will look at all the different risk elements within a business (here SME) credit application. These are: Risk of proprietors Risk of business Reason for loan Financial ratios Size of loan Risk industry Risk of region Before we delve into this list, it is worth noting that all of these factors need to be deployable as assessment tools within your originations system so it is key that you ensure your system can manage them. If you are on the look out for a loans origination system, then look no further than Principa’s AppSmart. If you are looking for a decision engine to manage your scorecards, policy rules and terms of business then take a look at our DecisionSmart business rules engine. AppSmart and DecisionSmart are part of Principa’s FinSmart Universe allowing for effective credit management across the customer life-cycle.  The different risk elements within a business credit application 1) Risk of proprietors For smaller organisations the risk of the business is inextricably linked to the financial well-being of the proprietors. How small is small? The rule of thumb is companies with up to two to three proprietors should have their proprietors assessed for risk too. This fits in with the SME segment. What data should be looked at? Generally in countries with mature credit bureaux, credit data is looked at including the score (there is normally a score cut-off) and then negative information such as the existence of judgements or defaults; these are typically used within policy rules. Those businesses with proprietors with excessive numbers of “negatives” may be disqualified from the loan application. Some credit bureaux offer a score of an individual based on the performance of all the businesses with which they are associated. This can also be useful in the credit risk assessment process. Another innovation being adopted internationally is the use of psychometrics in credit evaluation of the proprietors. To find out more about adopting credit scoring, read our blog on how to adopt credit scoring.   2) Risk of business The risk of the business should be managed through both scores and policy rules. Lenders will look at information such as the age of company, the experience of directors and the size of company etc. within a score. Alternatively, many lenders utilise the business score offered by credit bureaux. These scores are typically not as strong as consumer scores as the underlying data is limited and sometimes problematic. For example, large successful organisations may have judgements registered against their name which, unlike for consumers, is not necessarily a direct indication of the inability to service debt.   3) Reason for loan The reason for a loan is used more widely in business lending as opposed to unsecured consumer lending. Venture capital, working capital, invoice discounting and bridging finance are just some of many types of loan/facilities available and lenders need to equip themselves with the ability to manage each of these customer types whether it is within originations or collections. Prudent lenders venturing into the SME space for the first time often focus on one or two of these loan types and then expand later – as the operational implication for each type of loan is complex. 4) Financial ratios Financial ratios are core to commercial credit risk assessment. The main challenge here is to ensure that reliable financials are available from the customer. Small businesses may not be audited and thus the financials may be less trustworthy.   Financial ratios can be divided into four categories: Profitability Leverage Coverage Liquidity Profitability can be further divided into margin ratios and return ratios. Lenders are frequently interested in gross profit margins; this is normally explicit on the income statement. The EBIDTA margin and operating profit margins are also used as well as return ratios such as return on assets, return on equity and risk-adjusted-returns. Leverage ratios are useful to lenders as they reflect the portion of the business that is financed by debt. Lower leverage ratios indicate stability. Leverage ratios assessed often incorporate debt-to-asset, debt-to-equity and asset-to-equity. Coverage ratios indicate the coverage that income or assets provide for the servicing of debt or interest expenses. The higher the coverage ratio the better it is for the lender. Coverage ratios are worked out considering the loan/facility that is being applied for. Finally, liquidity ratios indicate the ability for a company to convert its assets into cash. There are a variety of ratios used here. The current ratio is simply the ratio of assets to liabilities. The quick ratio is the ability for the business to pay its current debts off with readily available assets. The higher the liquidity ratios the better. Ratios are used both within credit scorecards as well as within policy rules. You can read more about these ratios here. 5) Size of loan When assessing credit risk for a consumer, the risk of the consumer does not normally change with the change of loan amount or facility (subject to the consumer passing affordability criteria). With business loans, loan amounts can range quite dramatically, and the risk of the applicant is normally tied to the loan amount requested. The loan/facility amount will of course change the ratios (mentioned in the last section) which could affect a positive/negative outcome. The outcome of the loan application is usually directly linked to a loan amount and any marked change to this loan amount would change the risk profile of the application.   6) Risk of industry The risk of an industry in which the SME operates can have a strong deterministic relationship with the entity being able to service the debt. Some lenders use this and those who do not normally identify this as a missing element in their risk assessment process. The identification of industry is always important. If you are in manufacturing, but your clients are the mines, then you are perhaps better identified as operating in mining as opposed to manufacturing. Most lenders who assess industry, will periodically rule out certain industries and perhaps also incorporate industry within their scorecard. Others take a more scientific approach. In the graph below the performance of an industry is tracked for two years and then projected over the next 6 months; this is then compared to the country’s GDP. As the industry appears to track above the projected GDP, a positive outlook is given to this applicant and this may affect them favourably in the credit application.                   7) Risk of Region   The last area of assessment is risk of region. Of the seven, this one is used the least. Here businesses,  either on book or on the bureau, are assessed against their geo-code. Each geo-code is clustered, and the projected outlook is given as positive, static or negative. As with industry this can be used within the assessment process as a policy rule or within a scorecard.   Bringing the seven risk categories together in a risk assessment These seven risk assessment categories are all important in the risk assessment process. How you bring it all together is critical. If you would like to discuss your SME evaluation challenges or find out more about what we offer in credit management software (like AppSmart and DecisionSmart), get in touch with us here.

Collections Resilience post COVID-19 - part 2

Principa Decisions (Pty) L

Collections Resilience post COVID-19

Principa Decisions (Pty) L