Blog

Latest News and Updates

Evispot Articles: Case #1 – Detecting ​​Default ​​Payments ​​of ​​Credit ​​Card ​​Clients using Machine Learning

Hope you have had a great summer! While many have been lying on the beach and enjoyed vacation, Evispot has worked on interesting machine learning cases. 

During the upcoming weeks we will share our experiences in a serie of articles. The idea is to give you insights and a deeper understanding of machine learning in credit scoring. The articles follow Evispot´s development process going from raw data to actionable insights – and our hope is that you will share our view: 

Machine learning is the future in credit scoring

Case#1 –  Detecting ​​Default ​​Payments ​​of ​​Credit ​​Card ​​Clients using Machine Learning

The first article is about predicting default payments of credit card clients, using demographic data, credit data and history of payment bill statements of credit card clients. In this example, we use logistic regression to briefly demonstrate a classification problem, and how classification can be more useful than just predicting the outcome variable.

Follow the link to get to the article:
Case #1 – Detecting ​​Default ​​Payments ​​of ​​Credit ​​Card ​​Clients using Machine Learning 

If you have any questions or data you want to explore feel free to contact:

Sebastian.Larsson@evispot.ai, CEO & Co-founder Evispot

 

read more
Tomas SelldenEvispot Articles: Case #1 – Detecting ​​Default ​​Payments ​​of ​​Credit ​​Card ​​Clients using Machine Learning

It’s a Platform BOOM! PSD II drives platform banking

Platform Boom: This is the second article in the series of articles on the topic ”Payment Service Directive II (PSD II)”

Platform, is it the buzzword of the century? Perhaps, and even most likely. However, no spark, no fire. A buzz does not emerge from nothing. Some goes as far as to say that in 30 years, 50% of total S&P 500 revenues’ will stem from platform businesses¹. One should always be reluctant to the reliability of such impressive numbers in the future. But, the same bold prognosticator states that over the past 10 years, S&P 500 platform companies have increased net earnings 20 times as much as the overall average of S&P 500. The number of platform businesses are also growing. There is for sure some truth in the prediction.

Exponential growth of platform businesses among the S&P 500 Source: Business Collective, written by the CEO of Applico

Exponential growth of platform businesses among the S&P 500 Source: Business Collective, written by the CEO of Applico

PSD II is essentially the ultimate platform business model enabler in the financial industry. It will decentralise the power of the dominant large banks by making data easily portable and many services, like making transactions, accessible for third parties. In light of the impressive numbers of the profitability of platform businesses, there will be an obvious battle of becoming the platform in the future. This shows through the initiatives of creating a standard API for the PSD II to come.

The classics:

  • Uber, the world’s largest taxi company, owns no vehicles.
  • Facebook, the world’s most popular media owner, creates no content.
  • Alibaba, the most valuable retailer, has no inventory.
  • And Airbnb, the world’s largest accommodation provider, owns no real estate.

Next up?

Will we also add “Company X, the world’s largest bank, provides no credit”? Because with PSD II it is not necessary that the banks are the platform owners.

In the latest years the discussion has been circling around complying to the minimum requirements of PSD II. But the banks should not be overlooked! There are several banks, who have understood the value of the platform business. These actors are rather aiming for providing an API of all kinds of services within the bank. However, the ones outside the scope PSD II not necessary free of charge.

We have previously written about The Open Bank Project, who provides an Open Source API for bank integration. They are essentially building an App store for financial services. It is no bank, they are only building the platform. These solutions will grow with PSD II, and network effects will play out. When put in the hands of the bank’s customers, either companies or persons, a completely new industry of financial service will emerge.

Different Platform Business Models

Now, lets try to concretize the concept of platforms businesses. The two gentlemen Evan & Davids² define it as businesses that act as an intermediary and tie together two sets of different (but often related) actors. The two groups need each other and rely on the platform to act as an intermediator and facilitate transactions between them. In some sense, these platforms enable an exchange of value-creation that otherwise wouldn’t take place.

The companies we used as examples above have gone through extreme growth-phases. However, these powerful, positive growth dynamics makes monetization a complicated matter. The platform guru, Parker, means that monetizing an offering too early can be the death sentence for the platform since it creates friction in the growth. In contrast, not having a thought through plan for pricing makes you set for an unprofitable disaster.

So, how are companies attacking this problem?

One common method to overcome this hurdle is the strategy often called users first, monetisation later. This method were used by Instagram and Facebook in their early days. Meaning that the business focus on building a user base to a critical mass before initiating a monetisation. Another common tactic is taking a transaction fee of each intermediated service, thus taking a cut of the “whole” price.  Examples of this tactic would be Uber and Airbnb. The monetisation strategies of these platforms are many and are highly dependent of the which kind of service of product the platform intermediating.

It will be interesting to see which tactics will be the “winning-one” for the fintech platforms when PSD2 incepts. Keeping the monetisation strategy top-of-mind is important in order to leverage rapid growth when the bell rings.

The company behind the Open Source API is making money on an annual commercial license fee, maintenance and support. The App store’s success depends on an interdependent function of how many banks are integrated with the API (thus potential customers) and the number of apps. More banks will result more apps, and more apps will result in more banks.

Evispot provides the financial industry with the decisions they need in the future. We are developing an AI platform for the financial industry. AI on demand. Easily integrated. Secure.
Get in touch, let’s have a chat!

¹ https://businesscollective.com/5-reasons-entrepreneurs-should-take-advantage-of-the-platform-business-model/
² Evans, David S., Two-Sided Market Definition (2009). ABA Section of Antitrust Law, Market Definition in Antitrust: Theory and Case Studies, Forthcoming. Available at SSRN: https://ssrn.com/abstract=1396751

read more
Tomas SelldenIt’s a Platform BOOM! PSD II drives platform banking

Changes in the legal landscape catalyze transformation: PSD II

This is the first article in the series ”Payment Service Directive II (PSD II)”

No other industry is in such a rapid transformation as the financial industry. It is not only driven by new technology such as Artificial Intelligence (and all what that implies for credit risk and personalization of financial services), but also changes in the legal landscape such as the introduction of General Data Protection Regulation (GDPR) and Payment Service Directive II (PSD II) in 2018. In this series of articles we will shed light on the PSD II, the EU “Open Banking” initiative. Whereas the first PSD was induced by the rapid changes in the financial market, the second version, PSD II, is set to catalyze and force new rapid changes in the market.

There are many good articles of the implications of PSD II. Instead of providing one more article to the already crowded space, we simply recommend you to read one or both of the following articles: Banking Hub (short and easy) or Whitepaper by Deutsche Bank (detailed).

However…

Here’s a two- word summary: Financial collaboration

Here’s a two-sentence summary: With PSD II banks are obliged to offer the possibility for third-party payment services providers (TTPs) to integrate to banks infrastructure and data through an open APIs. This implies (with consent) at minimum accessing account data and being able to make transactions on behalf of the person.

PSD II enables you to pick your own personalised basket of financial services

PSD II enables you to pick your own personalised basket of financial services

Progress in the field

It is now only six months ahead of the implementation of PSD II (January 2018). From January banks have two years to adjust. The implementation is expected to come in four phases: starting with a minimum viable product (MVP) throughout 2018, and completed in Q1 2019¹. However, several large banks have already begun. Nordea who is paving the way in the Nordics is releasing a pilot test of their open API during the Autumn of 2017. Even though there are many initiatives to harmonize the API such as the recent announcement from the Berlin Group² or the German Open Bank Project, one can expect it to be many different APIs battling to become the standard³. There is an obvious first-mover advantage to begin the development early, i.e. to being able to control the formalisation of the one, the chosen, the API to rule them all.

There are numerous companies that have made a business out of having done multiple integrations with banks. Fidor Bank, Plaid, Instantor, Railsbank and Tink are examples of companies before their time. Throughout time with more harmonized APIs it will become easier to connect to many banks, but it won’t happen instantly.

PSD II and GDPR: A love story

It is an interesting combination the PSD II and the GDPR. Private persons have always owned their own financial data, but for the first time this really matters. Private persons have obtained a control position with the GDPR and the ability to move data easily with PSD II. In other words, banks and financial service providers will need to become much more customer-centric. Because for the first time in history one can swiftly change service provider and pick a basket of financial services from different actors.

We are welcoming the open banking of the future and the tectonic shifts in the legal landscape making it possible. In the next article we are interviewing another company who are actively working with PSD II. Stick around!

In the meanwhile, send us an email or give us a call to hear more about our plans with PSD II and the classification of transactions.


¹ http://docplayer.net/15301650-The-open-banking-standard-unlocking-the-potential-of-open-banking-to-improve-competition-efficiency-and-stimulate-innovation.html
² http://docs.wixstatic.com/ugd/c2914b_affaa594470e40e5ae16ced6b0e412d0.pdf
³ See also the dynamic list of open APIs in the financial sector: http://genome.dailyfintech.com/t/lets-build-a-list-of-banking-apis/201/33

read more
Tomas SelldenChanges in the legal landscape catalyze transformation: PSD II

… and the winners are! Evispot AI Challenge

Most Accurate Solution 


1. EZW (70,3%)
By focusing on the core problem EZW was able to developed a superior solution compared to the competition. EZW had a structured pre-processing and used random forest algorithm which resulted in the winning solution.
You now have the right to brag about your machine learning skills! 

A big congrats, well deserved! You are now going to tell Bisnode how they should implement machine learning.

2. Real Human Beings (69,1%)

Well done, Real Human Beings! A well deserved second place! Congratz. You will get azure passes!

3. Klubb Bubbel (64,7%)
Congratz! You will also get azure passes!

 

Most Innovative Solution 

Deep Engima
Many of the solutions followed a similar processes in terms of pre-processing and choice of algorithm – but Deep Enigma stood out in terms of application area. Deep Engima is winner of the most innovative solution – by proposing a suitable & creative application area of your solution! Congratz!

You are winning a night of beers with Evispot – let’s talk about how to develop your proposed solution!


Once again, a big thanks to Mikael Kågebäck, Christian Lauritzen and David Fendrich in the jury and to our partners! 

Congratulations to the winners! 

read more
Tomas Sellden… and the winners are! Evispot AI Challenge

Evispot AI Challenge: Powered by Bisnode – has now started!

Yesterday, Evispot kicked off the Evispot AI Challenge, a five day long machine learning competition, co-organized  with Bisnode and Microsoft. With just over 20 teams participating in the challenge Evispot invited the teams to a night of inspiring talks, food & drinks and introduction to the challenge itself. Besides Evispot, we listened to talks from David Fendrich of Crawlica under the theme “Why can we predict anything at all?” and Robert Quinn of IBM under the theme “Did you say Let’s eat Grandpa or Let’s eat, Granpa?”. Now the participants have until Tuesday to complete the case – we are looking forward for the results!

 


Photo: Fabian Wennerbeck 
Joachim Karlsson, Group Director of Innovation, Bisnode

Joachim introduced Bisnode and their view of the Evispot AI Challenge as well as highlighted the importance of machine learning in the future of credit decisions.

 

 


Photo: Fabian Wennerbeck 
”Why can we predict anything at all?” by David Fendrich, CTO Crawlica

David started off by saying that he would give a technical speech, not to talk about business at all. And he certainly did, David speech focused around the Solomonoff induction, Kolmogorov Complexity
and gave some valuable tips of how to tackle AI problems. You can find David’s slide here. 


Photo: Fabian Wennerbeck
“Did you say Let’s eat Grandpa or Let’s eat, Grandpa?” – by Robert Quinn, Senior Software Engineer, IBM

Robert talked about how IBM Watson cognitive journey started with playing Jeopardy. By building a Jeopardy AI-robot, with the two requirements, it should answer within one second and at least be 85% sure to answer the correct answer.


Photo: Fabian Wennerbeck 
Isak, CFO & Co-Founder, Evispot
Isak highlighted the importance of the challenge due to the upcoming regulatory changes in the financial industry.




Photo: Fabian Wennerbeck 
Tomas, CTO & Co-founder,  Evispot
Tomas introduced the challenge in details and declared the challenge’s start. 

 

Below are more pictures from the evening.


Photo: Fabian Wennerbeck 

Photo: Fabian Wennerbeck 

Photo: Fabian Wennerbeck 

A big thanks to everyone who showed up and had a great evening with us! If you have any questions or comments don’t hesitate to contact us – hello@evispot.ai 

Now we are looking forward for the results!

A special thanks to Fabian Wennerbeck who took the pictures!

read more
Tomas SelldenEvispot AI Challenge: Powered by Bisnode – has now started!

Credit Scoring on Anonymized Credit Data using Machine Learning

The financial industry has a tradition of data-driven decisions, but uses only a fraction of the available data. Financial companies generate vast volumes of data in their credit operations, which is scarcely feeded back into the operations. From an international perspective, financial companies in Sweden have access to good quality external data. But, there is much to by exploring machine learning to create new insights from the data generated by the credit operations. 

 

2018 – The Year of Regulations 
Apart from recent advancements in the field of machine learning in credit scoring and rating, 2018 will be a year of major regulatory changes. These regulatory changes, in combination and alone, will become highly important for the implementation of machine learning algorithms in the financial industry. The first regulation, PSD II, will make financial data available through standardised APIs for third parties allowed access by the owner of the data. The second regulation, GDPR, will strengthen the control of the owner of the data (only for physical persons) have on their personal data. The combination of giving stronger control to data owners and opening up the traditionally closed systems expose many opportunities for new entrants. With GDPR the use of personal data in business will be stricter, making anonymisation of data more relevant than ever. In the setting of anonymised data machine learning will play a crucial role to find patterns and anomalies useful for credit assessments.

 

The Evispot AI Challenge 
Together with Bisnode we are exploring next generation credit decisions by using machine learning on anonymised transaction data. We took it one step further by organising the Evispot AI Challenge, where we asked students of Chalmers University of Technology to help us find innovative solutions.

Joachim Karlsson, Group Director of Innovation at Bisnode, on the collaboration:
”Industry leaders are choosing Bisnode to help them to work better and innovative, to create growth and to find new opportunities using smart data. Our collaboration with Evispot is a way to let innovators take part of Bisnode smart data. Historically, we have positive experience in similar collaborations and we believe there are many good ideas in Evispot’s network and we want to support that.”  

 

This case reflects one out of many opportunities in using the internal data in credit decisions. The case sets the ground to make way for identifying purchase and payment behaviours to be used in the credit assessment of granting new debts.

Want to take part of the results and insights?  
Fill in your contact details in the following link and we will get back to you:
Follow this link

Do you have any comments, thoughts or questions around this theme?
Don’t hesitate to contact us on hello@evispot.ai or at twitter.

read more
Tomas SelldenCredit Scoring on Anonymized Credit Data using Machine Learning

Kvar-att-leva-på-kalkylen (KALP) är pånyttfödd!

För drygt tre veckor sedan annonserades det att H&Ms dom 529-16 om bristande kreditprövningar för konsumenter, nu tas upp i högsta instans. Detta beslut hade tidigare hävts i Kammarrätten den 19 oktober 2016. Fallet har nu pågått i snart tre år, sedan Konsumentverket (KO) startade utredningen för att motarbeta överskuldsättning i samhället. Om Konsumentverket får rätt i sak, kommer stora förändringar att följa för hela kreditbranschen, men för fakturakrediter och kontokrediter i handel och ehandel i synnerhet. KALP har fått en pånyttfödelse. 

KO menar att H&Ms kreditprövning har allvarliga brister i att förstå en konsuments totala betalningsutrymme. De sopar undan H&Ms argument om låga kreditförluster då de säger att det bara betyder att H&Ms kunder betalar just H&M. En kreditprövningen som tar hänsyn till konsumentens betalningsutrymme måste ha färskare inkomstsiffror, samt kostnader och andra skulder, hävdar KO och föreslår att kreditgivare måste samla in dessa uppgifter.

H&M svarar att det skulle försämra kassaupplevelsen, men även deras reviderade kreditpolicy blev avfärdad av KO. Motivering var att H&M använt KOs egna beräkningar för hushållskostnader som används i andra sammanhang.

Hur ser Evispot på KALP-kalkylen?

Vi på Evispot följer utvecklingen gällande KALP och vakar efter nya uppdateringar. I takt med att mer information släpps arbetar vi för att förstå hur vår AI kan bistå. Ett frö för en teknisk lösning är sått i Evi-LAB. Lösningen handlar om att låta kunden bekräfta föreslagen information istället för att fylla i allt från grunden. Genom att minska antalet steg för ett köp, ökar konverteringen. Det är allmänt känt. Ambitionen är således att skapa en smidigare kassaupplevelse.

Har du egna tankar eller är du intresserad av denna lösning är du varmt välkommen att kontakta oss på hello@evispot.com

(Sorry english speakers, this is regarding a Swedish regulation)

read more
Tomas SelldenKvar-att-leva-på-kalkylen (KALP) är pånyttfödd!

Extract Insights from Your Data

 

We are living in the data age. Every device around us is constantly assessing, calculating and interacting – resulting in enormous amounts of data constantly being created.

The term big data has been frequently used during the last decade. The term is used to describe the collection and analysis of data on a scale or of a complexity that makes the use of data challenging.  Extracting insights and value of these massive datasets is challenging, but when done properly you will find yourself increasingly efficient.

The UK house of commons wrote a study named The big data dilemma on the massive increase of data. The study highlights the importance of taking action and using the data to create valuable insights for your business. To add some perspective: in 2014 204 million emails were sent and every minute 4 million Google search queries were done. Even more astonishing is that 90% of the data currently in the world was created in the last two years and the amount of global data is predicted to grow 40% year on year for the next decade.

– So what can you do?                                     

The credit scoring giant Experian believes the role of big data in financial service is huge, naming fraud detection and general ledger data to gain previously impossible insights as some of many of possible products. Experian states that enabling  ‘real time’ data-decisions is the difference between winners and losers in the financial markets.

This data explosion offers a great opportunity of understanding your environment and peers on a deeper level. Properly exploited, this data should be transformative, increasing efficiency, unlocking new avenues in life-saving research and creating yet unimagined opportunities for innovation. However, the current datasets are nowhere near fully exploited despite research showing that data-driven companies being 10% more productive than those that do not operationalise on their data. Today an estimated 12% of all data analysed, meaning there is yet much to explore.

– Extracting insight and value from these massive data sets is difficult, but when done properly it can be the difference between winning and losing. Evispot was founded on the mission of being experts of insight-extraction in the data age.

Want to learn more about how your business can benefit from data?
Check out the Evi-LAB – our environment for insight extraction.

read more
Tomas SelldenExtract Insights from Your Data

Lansering av Evi-Lab

Evispot.AI lanserar den intelligenta testmiljön Evi-LAB som hjälper finansbranschen att ta smartare och enklare beslut

Tomas Selldén, Isak Unfors och Sebastian Larsson

Evispot.AI, backat av Chalmers Ventures och Invativa AB tar sikte mot att utveckla nästa generations kreditbeslut. I utvecklingen av en AI-beslutsmotor för finansbranschen, lanserar de nu testmiljön Evi-LAB. Syftet är att minska tröskeln för att skapa smarta och enklare beslutssystem för framförallt kreditgivning och kredithantering genom initiala tester i en säker labbmiljö.

Till skillnad från dagens system så ska Evispots modeller optimeras och uppdateras i realtid. De stora volymer finansiell data som idag genereras vid kreditgivning ger upphov till en enorm komplexitet som varken människa eller konfigurerad mjukvara kan hantera, desto mindre i realtid. Grundarna av Evispot bestämde sig för att underlätta för finansbranschen genom att utvinna information som gömmer sig i de stora mängder data med AI och göra den tolkningsbar.

”Det är viktigt att jobba systematiskt när det kommer till att utveckla smarta och automatiserade beslutssystem. Evi-LAB handlar om att validera möjligheterna från faktiskt data. Det handlar om att bygga från evidens” – säger Sebastian Larsson, medgrundare och VD.

”Under 2018 kommer två lagar med stora implikationer för finansbranschen att implementeras, PSD II och dataskyddsförordningen. Detta kommer att förändra hur och vem som får användning av denna data. Vi positionerar Evispot långsiktigt i de snabba rörelserna, med nya samarbeten och den senaste teknologin för kreditbeslut” – säger Johan Kohlström, medgrundare och grundare av Invativa AB.

Gå in på www.evispot.ai för ett evidenstest i Evi-LAB eller spana in projekten som är på gång.

Kontakt:
Sebastian Larsson, Medgrundare och VD Evispot.AI
sebastian.larsson@evispot.ai
070-233 23 35

read more
Tomas SelldenLansering av Evi-Lab