Device and method for analyzing reputation for objects by data mining

This invention describes a software system designed to assess the reputation of specific items or entities by analyzing information found on web pages. It works by first gathering data from the internet, then breaking down the text into its smallest meaningful units (morphemes) to understand its content. It then categorizes this information, identifies the specific items being discussed, and determines how sensitive or risky the information is based on certain keywords, ultimately calculating a numerical risk score. The claims specify that the system includes a news crawler, an HTML parser, and a spam filter.

Why it matters: Filed before the widespread adoption of advanced AI and large language models, the complex text analysis, morpheme processing, entity recognition, and sensitivity classification described in the claims can now be performed with significantly higher accuracy and efficiency using modern NLP techniques. What required extensive custom engineering in 2015 is now more readily achievable with off-the-shelf AI tools.

Status: Not in forceListed as no longer active. The specific reason is not in the record we hold.
How hard to build: SpecializedHeavy software system, NLP, data mining logic

3 AI reinvention directions for this patent

See all three

AI gives you a few directions you could take this. Pick one, and we check whether your version is different enough to patent, then write the filing.

Reinvent this with AI

Where this one came from

Every morning we go through millions of abandoned patents and pick out 15 worth your time. This was one of today’s.

This one

+10