With the help of TEXTA Toolkit and support from our analysts,
our partners can solve most relevant tasks related to text analytics.

Document Searching img
Text analysis solution img

CUSTOMER SUPPORT
AUTOMATION

Information extraction img

INFORMATION
EXTRACTION

Document audit img

DMS AUDIT

Software to find similar documents img

DOCUMENT
RECOMMENDATION

OCR img

DETECT TEXT
FROM IMAGES

Document classification img

DOCUMENT
CLASSIFICATION

Sentiment analysis img

SENTIMENT
ANALYSIS

Statistical information img

DATA
VISUALIZATION

WITH TEXTA TOOLKIT YOU:

Save in labor cost

Understand better your products and performance of different units

React quicker to business events

Gain control over your DMS

Less routine tasks for your workforce

PRICE:

TEXTA Toolkit is open source and freely available.

PARTNERS

THE MINISTRY OF EDUCATION AND RESEARCH

Document Management Leak Analysis.

The Ministry of Education uses TEXTA Toolkit and it’s extensions to audit document management in order to identify documents which have gone public without permission (e.g. health documents or work contracts). As a result of the initial project, the documents errantly published were identified by our data scientists and the leak was closed.

CV ONLINE

Document Recommendation.

CV Online is using our recommendation engine to find job ads and CV-s suggested to the user while browsing the site. We have combined information extraction with unsupervised machine learning and optical character recognition (OCR) to build a recommendation engine capable of processing and suggesting documents in various formats including images and scanned documents.

EKSPRESS MEEDIA

Hate Speech Detection.

One of the largest media groups in Estonia - Ekspress Meedia - is using TEXTA Toolkit and our machine learning models to automatically identify and remove toxic or violent content from the commentaries of online media.

ÕHTULEHT

Topic Tagging for Newspapers.

Estonian newspaper Õhtuleht is using TEXTA Toolkit and our machine learning models to automatically tag their published articles with keywords and entities (e.g. names and locations) describing the text. We have witnessed the AI-based algorithm producing the keywords more coherently and homogeneously than it’s human counterpart, thus improving the navigability of the archive.

ESTONIAN RESCUE BOARD

Text Mining for Resource Planning.

TEXTA Toolkit is used to extract information from dictates issued by the Estonian Rescue Board in order to better channel their inspection activities based on data analysis.

CENTRE OF REGISTERS AND INFORMATION SYSTEMS & MINISTRY OF JUSTICE

Automatic De-identification of Judicial Decisions.

Using our software, the Ministry of Justice in cooperation with the Centre of Registers and Information Systems stripped personal data from nearly 80,000 judicial decisions involving expunged punishments and thereafter made the decisions available in the court information system again.

This project received an award for the best IT project of 2019 in the Ministry of Justice: https://www.facebook.com/permalink.php?story_fbid=2295147893930969&id=115173698595077

INFOREGISTER

Information extraction from court decisions.

Output of the project is an analysis engine of judicial decisions, which regularly processes the documents of the registry of judicial decisions and identifies information describing the result of lawsuits, which will be then readily made available for the customer. To solve the task, we used partitioning of information to find relevant information from the text – who were the parties to the lawsuit and how did the legal process end.

CONSUMER PROTECTION AND TECHNICAL REGULATORY AUTHORITY

Topic Tagging for Customer Support.

Using TEXTA Toolkit, our data scientists analysed the email communication of Consumer Protection and Technical Regulatory Board to build machine learning models for automatically tagging incoming customer emails. As a result of the project the tagging process of emails was automated and several bottlenecks in communication processes were identified.

NATIONAL LIBRARY OF ESTONIA

Topic Tagging for Libraries.

The National Library of Estonia is using TEXTA Toolkit coupled with custom modules to automatically tag various types of publications (e.g. books, dissertations, and articles) with keywords and entities from Estonian Subject Thesaurus (https://ems.elnet.ee). This solution is aimed to speed up the cataloguing process of new publications.
OTHER PARTNERS
aripaev
stacc
RIA
Scoro
CreditInfo