At Siren one of our major focuses is on investigative intelligence. The Siren platform merges functionalities that were previously disconnected, such as big data dashboards, link analysis, search engines, and operational monitoring. It extends data analysis with unique associative and multimodal analytic capabilities, finding relationships between data sets that were simply not visible before.
Image attribution: Siren.io
There are many Business Intelligence (BI) platforms in the market and, undoubtedly, they do provide value, however, as a rule, we tend not to compete with them. In general, BI solutions are relevant when you have a specific use case, for example if you’re trying to analyse sales revenues by territory over a month or year. This is a clearly defined requirement; it might be something you look at every day – but it’s always the same thing. Data is modelled for this use case and reports are created using the BI solution. We can do this, but it’s not our main focus.
SEE ALSO: Next generation of games: the importance of data for instant experiences
For investigation, you do need some BI capability, but you need to be able to quickly change the questions you’re asking. In the world of investigation work, there are individuals whose line of questioning is unknown at the outset. The investigator must follow leads and ask complex follow-on questions based on the answer to the last one. Consider an intelligence analyst exploring a terrorist cell; a police detective looking for organized crime links to automobile thefts; a fraud analyst looking for anomalies in payments – all of these individuals share a common theme. The unknown.
Using the sales analogy, perhaps you want to know why a number has dropped in the last month. This means understanding your usual events and determining change, perhaps suppliers in the chain have changed their terms…it is all about shifting perspectives. Without requiring heavy re-engineering or coding, Siren enables analysts to accommodate change and quickly get insight.
We don’t replace the business knowledge the users have, what we do is increase productivity and give the user more options. Part of my role is to deliver Proof of Concepts (PoC). In a recent PoC, I worked with a prospect who spent a lot of time integrating different datasets, which could, for example, be data from recently acquired companies – the challenge for the analyst is that data very rarely aligns, will be in various formats and will be categorised differently. This means analysts spend days trying to unify data from multiple groups so that they can understand what data they have, generate reports, assess performance and ensure compliance.
We have a fully automated functionality which identifies relationships across different datasets, and in this PoC, the analysts found it very empowering. For example, if you’re working with different organisations, categorisations vary. In one company you might be evaluating “customer account”, but in another company, the same thing might be called “account identification”. Using Siren you’re able to recognise that these two different concepts are the same thing. This is a big leap in understanding interconnected data.
When you have to start analysing new data, we automate a lot of this process – making it more streamlined which facilitates the interaction with the user. Many users might spend a day trying to understand data, using Siren you can start to see results almost immediately and make sense out of it.
We increase productivity and also see gaps an analyst might not normally see. What do I mean by gaps? Relationships often exist across data, but as an individual, you can’t see them. Siren has the functionality to automatically find relationships across data – avoiding missed opportunities.
Image attribution: Siren.io
When we undertook training for this prospect and users started interacting with the platform, they soon realised they could quite easily identify interconnected data, but also profile it. For example, datasets might exist which the analysts aren’t familiar with.
What Siren does is create finger-printing, so when you import new data you can automatically identify its type, what records of the data exist, if it’s a number we can identify the range, if it’s a date we can define time periods. We can also quickly generate an automatic dashboard – this is helpful as when you’re creating a dashboard from scratch, if you don’t understand the data, you don’t really know how to visualise it. By using Siren what could normally take a couple of hours is generated in a couple of minutes.
SEE ALSO: What data should AI be trained on to avoid bias?
Image attribution: Siren.io
In the past I have demonstrated Siren with no data and loaded three Excel spreadsheets into the platform and created dashboards in about 20 minutes. This goes from having nothing but spreadsheet data to having integrated and fully functional dashboards in a very short window. Using relationship discovery and automatically creating dashboards, the basis for analysis can be achieved in a couple of minutes, even using Siren’s community edition. If you’re in a situation where you have to make an informed decision quickly, Siren can really help.
The post Using Siren to increase personal productivity and enable quick insights appeared first on JAXenter.
Source : JAXenter
