THE PROVIDENCE ADVANTAGE
Collect data from social media, blogs, forums and the darknet
Data Collection: Data collected by Providence includes usernames, posts, images and post metadata and is refreshed every 30 seconds
- Over 130,000 news sources
Data Fusion: Correlate events across platforms, thanks to our proprietary data schema which fuses unstructured data feeds
Bulk Upload: Set up an intelligence operation quickly by uploading entity and interest data in bulk from spreadsheets
Evidence Support: Data is captured and stored exactly as it appeared on the data source, so it can be used as evidence in investigations
Persistence: collected data is stored in the cloud so it is available to users even if the original content is deleted from the data source
Identify events of interest, quickly
Dashboard: Quickly identify concepts and information of interest and drill down into the details
Data Visualisations: The dashboard is comprised of configurable data visualisations including;
- Graphs and charts such as pie charts, activity graphs, bar charts and many others
- Geolocation mapping
- Word cloud
- Link charts
- Data tables
All visualisations are interactive allowing the user to filter very quickly to specific information of interest
Search: Search within collected data. Use Boolean search and regular expressions to filter data shown on the dashboard and find information of interest quicker
Relevancy Ranking: Tackle your most important events first. Based on your past analysis, the relevancy engine uses machine learning and natural language processing to automatically prioritise the events most likely to be of interest
Share intelligence across platforms and with other teams
Reports: create PDF reports of the data you have collected and the events you have identified
Scheduled Reporting: Receive reports based on your schedule, so you see the data you need when you want it
Alerting: Be alerted via email or SMS when specific events or topics are mentioned
Data Export: Download the data collected by Providence into .csv or JSON formats for use in other applications or for record keeping purposes
A Secure Platform
Secure Hosting: Providence is hosted in the Amazon4 Web Services (AWS) cloud in Sydney. The AWS infrastructure is approved by Australian Signals Directorate (ASD) and is on the Certified Cloud Services List (CCSL). Each customer has its own instance of Providence deployed within a virtual private cloud
Anonymous: Providence is deployed in a way that is nonattributable to both WorldStack and our clients. All API keys used for collection and IP addresses used by Providence are generic with no links to WorldStack or our clients
NOT SURE WHAT OSINT 2.0 IS?
OSINT 2.0 is the next generation of OSINT services that include artificial intelligence technologies such as machine learning and natural language processing.
The first generation of OSINT services focused on automating the collection of data from online platforms including social media, forums, blogs, and news sites. It also included techniques like sentiment analysis, but not the same sophisticated, adaptable AI-driven technology we use today.
Social media monitoring is a subset of OSINT services that focuses only on social media platforms. Sentiment analysis is a service, typically text based, which aims to derive the intention behind a document, and is not specific to any platform. In first generation OSINT services, this sentiment analysis engine is typically rules-based rather than driven by machine learning techniques.
OSINT 2.0 services build on these first-generation services by leveraging artificial intelligence techniques to automate the intelligence process after collection. Examples of these services include automated triage based on Analyst feedback, discovery of online accounts, sentiment analysis, and risk-scoring of user behaviour.
FREQUENTLY ASKED QUESTIONS – PROVIDENCE
Not at this time. We support searching in multiple languages, but translation of collected content is not yet available. We are planning on implementing this feature soon. Manual translation is available via use of browser plugins.
The keyword number associated with the team and enterprise plans is guidance based on the relationship of keywords to average number of posts. Some keywords result in a lot more posts than others, depending on the popularity of the keyword. When the instance reaches 85% capacity, the customer will have a choice to archive the data by exporting it to file, deleting data or upgrading their plan to add additional capacity.
OUR CASE STUDIES
DO YOU HAVE THE COMPLETE PICTURE
Providence uses Machine Learning to automate the process of finding information of importance for users. It finds this information much faster and more accurately than any user can do manually. It also automates the relevancy ranking (Low, Medium, High) to enable better and more immediate prioritisation for users.
As a case study, we took two weeks of social media posts on cyber security (1.6 million posts) and manually applied detailed filtering. We found 400 posts of relevance in the 1.6 million (of varying relevancy ratings). Providence was then applied to the entire 1.6m using 200 of the 400 posts we found as training data.
Not only did it find the 200 withheld, it found a further 200 (600 posts in total) and accurately applied relevancy ratings to them.
The 1.6 million posts represented 2-weeks of data and included 45 high-relevancy posts, translating to an average of approx. 3 High-Priority posts per day. Not only do Analysts receive important information much faster, the likelihood of missing critical data is substantially reduced.