Watchman Analytics™ Optional Components are designed by Intuidex to operate in multiple environments, whether in existing agency system information process flows, or the Watchman Analytics™ RIA Framework. The optional components can perform on-demand in visual analytics frameworks, or operate independently as embedded analytics in enterprise data process flows to improve the quality and quantity of data for all downstream users.
Watchman Analytics™ can continuously monitor near-real-time data streams and provide alerts based on user defined parameters and cross schema data analysis. For example, free-text analysis of army chatter, combined with other intelligence sources, could identify the location of a hostile party assembling at a known vantage point within view of a protected area. Alerts may be sent to one or more recipients through established channels (e.g., email, SMS), or to other applications via Web services, and to third-Party alerting applications.
IxRules™ is a semi-supervised, active learner for information extraction rules. IxRules™ supports efficient rule composition through semi-automatic discovery and creation through an iterative process of user input and system feedback. The resulting rules are set in standard format usable in engines available from Intuidex and other vendors. IxRules™ reduces the time and resources otherwise required for information extraction, and is used with IxExtract™ to deploy enterprise-wide entity extraction.
IxExtract™ is an information extraction solution that returns input text enhanced by the extracted recognized entity values and types. Built featuring the Intuidex IxEEE™ engine, IxExtract™ helps to structure unstructured text, allowing for richer analytic results from incident narratives, email reports, and large database fields. It is used for enhancing data being sent to shared repositories, processing of generated products and reports for linking in archives, and is accessible as an always-on stateless Web service.
IxClassify™ incorporates Intuidex’s Higher Order Learning™ technology to assess the relevance, urgency or importance of various kinds of data. Lab tests have shown the value of Higher Order Learning™ on ecommerce data, SIGINT data in defense applications, nuclear detection data homeland security applications, and unstructured textual data in a number of applications. For example, given a stream of “be on the lookout” (BOLO) alerts, IxClassify™ can automatically assign a threat level for each document, a feature enabling users to focus attention on the most critical alerts, and also forward informative BOLOs to a secondary data processing path for further analysis. IxClassifyTM supports data processing workflows by categorizing or filtering SARs, BOLOs and other dynamic data streams, to identify and highlight content most relevant to the information need.
IxClassify™ scales to any number of different categorization or classification problems, and will function in near-real-time to categorize events in both law enforcement sensitive and open-source data in response to user queries.
IxForecast™ is designed as a data trending solution supporting analysis of data sets over time for prediction and forecast modeling. After users provide a time-series of data, the component generates a new time-series set describing future data points which follow the trend ‘learned’ through analysis of the input. IxForecast™ supports several forecasting models for different kinds of data (e.g., numerical, geospatial), and allows users to specify the parameters of the forecasted dataset to be returned (e.g., future time endpoint, number of future values). When used in concert with other components, users can specify parameters in a visual client, and display the entire time-series (existing and forecast) using Watchman Analytics™ ready chart and map components.
Higher Order Cohort Detection (IxCohort™)
IXCohort™ was developed in partnership with our law enforcement partners and leverages Intuidex’s Higher Order Learning technology on relational databases and related unstructured data to determine higher-order “cohorts,” or sets of items such as people, places and things that are related.
Intuidex's IxRecommend™ recommender system gives user's timely, relevant recommendations of interesting information that meets their financial, business, technical, scientific and other news information needs. IxRecommend™ considers a user's interests and makes decisions about what to recommend on-the-fly using sophisticated data analytics technologies including Intuidex's IxHOL™ Higher Order Learning™ technology. IxRecommend™ operates seamlessly and efficiently as an integrated component able to meet virtually any website's recommendation needs.
Custom Reports: Watchman Analytics™
Custom Reports: Watchman Analytics™ supports report generation of analytic results directly into a number of proprietary systems (UCR, CompStat).