Enter the content which will be displayed in sticky bar


Subscription Analyzers: Turnkey Text Analysis

All Code is 100% Transparent, Flexible, with Regular Updates


From entity extraction (companies, people, products), to sentiment (what are people saying about your products or a client’s products), to processing resumes, to extracting contacts, emails, etc. – we have turnkey analyzers for you.


To break down text to its syntactic components, you need TAIParse. It breaks each sentence down to its simple clauses, constituents such as noun phrases, verb phrases, and prepositional phrases, all the way down to parts of speech such as nouns, verbs, adjectives, and adverbs. TAIParse provides the backbone for customizing analyzers to extract the content, meaning, and information from text.

Formatting & OCR

If you are spidering websites, chances are the webpages will be constantly changing. NLP++ is a great language for parsing HTML pages and focusing only on the information you need. NLP++ also includes specializations for OCR cleanup, as well as handling the free text in documents with any degree and type of formatting.

Entity Extraction: People, Companies, and Products

What is It About?

To find the major players and products in a text, you need entity extraction. It finds the names, companies, products, and titles in a text, even pointing out names that have not been pre-wired into the system. Using context and a body of current knowledge, the entity extractor identifies the “who, what, where, and when” in a text.


Resume Analysis: Off-The-Shelf


Getting the Facts

Resumes come in myriad formats that are constantly evolving. Enter TAI’s Resume processor. Why reinvent the wheel and write your own when you can simply subscribe to our resume processor and get regular updates.

Sentiment Analyzer: How do People Feel?

Reading Between the Lines

With so much unstructured text out there in Tweets, Facebook, forums, and the like, there is invaluable information to be had for you and your clients. TAI’s sentiment analyzer can help you identify people’s impressions of your clients, their products, and employees. Terms can be added on the fly to enhance the analyzer without reloading the engine.


Business Events: Know Who’s Doing What


Finding the Action in Business

Business intelligence is a critical part of modern commerce. Which company is buying which, who hired whom, what products have just been launched, who got listed on which exchange… Such news can be automatically processed by our Business Events Analyzer to help you track and predict what’s happening in the marketplace.

TAI Parse: Diagramming Sentences

The Structure of Free Text

Much like diagramming sentences in school, TAIParse can automatically determine the structure of text. That backbone enables domain-specific knowledge (or “semantics”) to be layered on top, so as to automatically extract your mission-critical information from the free-form text.


Official Records: Document Infromation Extraction


Real Estate, Court Documents, Title Plant, Clerk of Court

TAI’s Official Records analyzers — via our XIEO partnership — can read documents that have been OCRed from images or paper and retrieve the information needed by real estate companies, title companies, and clerk of court. The XIEO team has decades of experience in these domains, from web scraping to image OCR to lights out processing, redaction, extraction, party indexing, property matchup, legal description analysis, and manual quality review, correction, and completion of the data to near-100%!

OCR Correction: Garbled Text

When Things Get Skewed

When documents come out of an OCR engine, watch out! Things don’t always line up and text gets glommed together and misspelled. TAI’s OCR cleanup can cleanly extract information from such messy texts. The analyzer takes over where OCR engines leave off, correcting zoning errors (such as tables that have been incorrectly split apart), spelling errors, numbers versus letters (“zero” versus “oh”, “one” versus “lowercase L”), and many other systematic and special-case OCR problems.