What Universal Challenge Do All Governments Face Today?
- Intelligence agencies require reliable identity resolution to fight terrorism and protect citizens.
- Border-control officers need accurate identity matches to fairly enforce immigration policies.
- Social service agencies with disparate databases must reconcile client identity, to both ensure benefits go to eligible people, and to eliminate fraud.
- Law enforcement seeks predictive capabilities to protect civilians from the potential danger of violence during demonstrations and periods of civil unrest.
- Financial regulators need accurate screening for public figures and financial institutions against watch lists across geographic, cultural, and national boundaries.
Though each of these challenges appears unique, all can be addressed using the same solution: trustworthy text analytics.
With data flowing with ever greater volume, velocity, and variety, increasingly your success is dependent upon:
- Identity verification
- Comprehensive data mining
- Immediate strategic insights
Ten Questions to Ask About Text Analytics
- Does the product analyze both structured and unstructured text?
Comprehensive text analysis should accommodate documents, web pages, databases, eDiscovery (i.e., legal documents), resumes, and all social media.
- Does the product perform accurately with any text source?
The most effective product will not only perform out of the box but will also be adaptable and expand to incorporate any formats specific to your field.
- How many languages are covered? The most reliable results can only come through broad multi-language capability. Ask particularly for expertise in both challenging languages (i.e. Arabic, Chinese, Korean, Russian) and languages spoken in conflict zones.
- How thoroughly does the product verify identity?
You want a product able to identify individuals through their legal name, nickname, initials, and possible misspellings, regardless of language.
- Does this solution rate the reliability of each identity match?
The most useful identity matching and verification includes a probability rating.
- If the product encounters the word “Clinton” can it differentiate between Bill and Hillary?
Your solution needs to interpret context so it can connect all words to the real-world people and things they represent, no matter the source.
- If multiple products are required, how well do they work together?
A suite of products designed to seamlessly integrate is more likely to provide greater synergy and consistency, and higher quality analytics than if you combine standalone products.
- Is the product customizable and scalable?
The most responsive and cost-effective solution will meet your immediate needs, add additional features as required; grow with your future needs; and can move to the cloud.
- Can the product be integrated into my existing infrastructure?
Look for text analytics that can be layered onto existing infrastructure and major search platforms, regardless of their being open source, legacy, secure or cloud based.
- Is text analysis a core competence of the provider?
When text analytics is a primary focus of a company, the product is continually developing, improving, and expanding its capability.
Be prepared – Take the time to assess if your current solution is as reliable as it needs to be.
Make sure it delivers the fastest and most accurate results possible, while minimizing errors.
Consider whether your solution offers the inherent scalability you need as your requirements continue to expand.