7 Reasons Why Effective Compliance Management Is Only Possible with a Graph-Based Solution Part112/20/2018 The real world is a graph and it's messy so we don't have simple models. Our data structures are very complex. They're always changing. Networks are changing.
Shareholders are changing. So we need to account for that. Also in this kind of field of work we often see the unexpected so more of the kind of creative business constructs you wouldn't usually use or find so we have to deal with that. We have to prepare for that. And also that obviously our work is framed very much by regulation. Regulatory change also is a constant. So we need to account for that. We need a flexible, flexible system to actually be able to deal with all of this and I think at the left hand side we see here is the networks from one of our production servers. So they are very complex. We need the flexibility within the data model. We need them with, for changing the data model. We also need to be auditable. So because it's compliance after all. So I can't just change everything. I also need to document it and make it comprehensible to a third party. Another important aspect is that we have high complexity. As seen by those networks and while I want this high complexity obviously in my backend because I wanna analyze it, I kinda need to condense it down to a level where actually users, clients or regulatory bodies can actually work with it. Because no one can actually see what's happening here. So I need to have, you need to kinda able to condense it down. And lastly also we must be transparent because our results and their processes they must be verifiable and again we're doing compliance. I need to prove that I did a good job. So unsurprisingly our main requirements is that we need to be able to store large amounts of complex network data efficiently. Be able to query it efficiently, flexibly. We need to be able to have a system that is very adaptable so we can actually react to changes. Whether they be regulatory or otherwise. Again we need to have allow complexity but also be able to dial it down a bit. Because yes we are an IT company but a lot of our users are lawyers. They also need to understand it. They are not graph experts. We also have to deal with complex security models. Because it's very sensitive data. Not everybody is supposed to see every bit of data. That is also a very important requirement to us and lastly I think providing secure storage and archiving it's kind of logic because what I need to show for at least the last five years that I did my job correctly. And lastly obviously it needs to be easy to maintain. Why all of these requirements are a perfect fit for Neo4j and especially Structr? I think this Axel will elaborate upon. - Thanks Julian. Yeah so now I have 10 minutes left to explain what Julian meant by these two products are the perfect fit for it. So I could simply say we fulfill all these requirements and we can stop here. But I have some more details to share. First technology decisions always important because sometimes it's difficult to change that afterwards once you have imported lots of data it's very difficult to change, your most basic layer, your database. So we chose the best database for connected data. Graph-shaped data as we all just saw it here. And the decision is very simple because there's the best database the best native database is Neo4j. And so we built, it is built and designed for native data for graph data. No other database is optimized for this use case. And furthermore it's stable and it's great for enterprise requirements. Now what is Structr here in this game? So Strutr is the application platform that helps to connect the data to the front end, the users use or the interfaces you import and export data with. As Julian pointed out we needed a high level of flexibility in almost any area that helped to reshape the data model. Or made it possible to reshape the data model at any time, at any given time. Even at run time because you cannot wait for development cycles. Like couple of weeks or even month to rebuild your application only because your network, shape of your network has changed. Or your regulatory requirements have changed. So the secret behind Structr is that the entire definition of the application is also a graph and stored in the same very same database in Neo4j. And you can manage everything in a single integrated tool which comes with lots of integrated and built-in functionality like the schema editor. Where you can change your data model according to the actual size of your data. And you can follow those changes very quickly. And it has some very handy components like an integrated content management system and a page builder, and a document store and file storage and so on. And the latest addition to this platform is our flow engine with a flow editor which is managed through a flow editor. That allows you to define queries, processes, logic of almost any kind, decision trees. Check routines and so on. In a graph interface. Or with a graph interface. So by dragging and dropping and connecting nodes and relationships which is very natural if you're in the graph space. You create these checks. My colleagues just talked about like the UBO check, like the identity status check. And you can reuse these components. The result of these components and connect it or use it as one node in a more, in a different check routine. So this is the ultimate tool to allow non-developers to create or maintain or manage logic which is very important here. To keep up with all these changes. So an effective graph solution only works if you can evolve as quickly as your legislation or your changes in your regular, not only in the legislation but also the changes your users need in the user interface. Cause everything is non-uniform and inconsistent. You really have to keep track on that. Very quickly. It comes with a broad spectrum of import and export interfaces and because most of the functionality, the basic functionality is already built-in to that platform. It's very cost effective which is very important if you want to scale. And the solution have to scale from a single person business up to an enterprise or a large company. It has of course to be secure and meet a high standard of security. Yeah so this solution we built really supports domain experts in their daily work by giving them the tool they need to define the processes without need from us developers. And that's I think a major step in this industry to really provide a tool that enables subject matter experts to do their work without being without having to rely on programmers. Well coming to a conclusion it's a very complex area with complex regulations. You need a smart software solution of course it cannot be done without a software. To know what a business or your customer means to you means that you have to know their graph by the way we're using here the same technology as the journalists by the, of the ICIJ use for the Panama papers and Paradise papers, West Africa leaks. So it's kind of natural to use this kind of technology in this area and so the technical components here are the native graph database and the very flexible application platform. So here are the seven reasons we promised to give you and the title. Money laundering happens in a graph. The world of money laundering is a graph. That's obviously one important reason. The competition is also using them like maybe it's not competition but it's a maybe all the different sides of this journalists and maybe even the people who launder money, I dunno. Maybe they are also using graphs we don't know. But if they do we can fight back using graph technology. The regulatory requirements can be best understood and also checked in the graph. We have a transaction graph if you connect the transactions of financial transactions of the people and the businesses to their nodes in the graph then you get an even larger graph. The security system is also a graph. You have a hierarchy of groups and roles and people. So these are important reasons and technical components also have to be a graph based solution like you need fast queries to do real time status reports. You don't want to calculate them overnight. And redo them overnight again if something changes in the graph you want to have that in real time that's only possible in the graph database. And to be flexible enough to keep track with all the pressure coming from your users, they want maybe new tools every week. You have to store also the application components, the UI components in the graph. That's what we do in Structr. And in the last step manage the business logic as a graph what we are doing here as well. So, that was basically it. Just some announcements we just released our new version 3.0. Yesterday there's a Windows installer coming with it so you can easily download and try it out. We have completely revamped our website. Yeah, question. - [Audience Member] Can you briefly describe what your platform does or what your application platform does? - Maybe we can, we have a booth outside just at that side you will run into our my colleagues and we can explain you everything about the platform. Yeah so if you want to follow us on Twitter and it's just @structr and the contacts are my colleague Inès she's also here in the audience and Christian from KERBOEROS Compliance and they will be able to answer you any question. Any questions if we have some time left? I dunno? Yeah? - [Audience Member] So, have you actually implemented transaction monitoring assistance from banks or is it more KYC based? - Yeah the question is have do we have implemented transaction monitoring for in banks? Or is it rather KYC? So KYC is a very important component which is already in place and we are currently as we speak adding transaction monitoring and transaction-- - [Audience Member] Case management. - Case management to the solution. - We have currently implemented the case management already within Structr within our application and we are currently transitioning from the previous transaction monitoring solution to implementing that into Structr as well. - [Axel] More questions here? - [Audience Member] How do you handle the time element? Cause you talked about it, you know, earlier how things change over time like the Structr that existed a year ago doesn't exist now? - Yup very good question. Question was how do we handle time management or changing things in the graph over time? Very good question we had a very fruitful and lengthy discussions about how to handle something like it's not, versioning of because it depends on the type of data in your graph. So what you can easily do is if you have a new version of an ID document or something like that or a contract you can just add it to a linear version, list. But if for example things change like who is owner of a business and this is more difficult. And we handle that by, I think we handle that best by creating new documents because for each new relationship a new document as proof or as documentation is added to the graph and then we can make the, we can add the information from when to when. Or until when was this document the valid document. And when the situation, when has the situation changed. You can explain it much better I think. - Well not from a technical point of view but yeah I think that was actually one of the bigger challenges and I think we forgot it up there. I'm sorry. What we actually do is yes we take the a document like a shareholders list. We connect it between all the nodes but what we also do is we create another relationship if it's a first time we upload that shareholders list for that particular company. So we create another relationship for each of the shareholders that is linked to the document as well. That is then written on by each additional document. So documents are classified so I know that if another document from the type of shareholder information is added to the graph it will write forth those relationships and is also added to those relationships as proof when these relationships existed. And obviously when a shareholder is not on any shareholders list anymore then this relationship will be ended, with a timestamp so we have all the past relationships we can still query them.
0 Comments
Leave a Reply. |
AuthorBen Goward is a founder of this Career Blog. He is a n Economist, business owner, successful businessman and contributor at Craftresumes. He spends his free time with his family, friends and favorite cat Mickey. Archives
December 2018
CategoriesWebsites we like:
|