01.10.26 (inclusive)

Стажировка в компании Kompra на 1 мес.

MVP

ICT tasks

Media sphere

Information processing and transformation

Software/ IS,

API/A reference book, etc.

In open data, the company is represented by its legal name and BIN, while in the media, social networks, and advertising it appears under a brand or trade name — and they often do not match. One legal entity may have several brands, the names are written in three languages (KZ/RU/EN) and in transliteration, with different organizational and legal forms (LLP, AO, IP), abbreviations, typos and spelling variants. Because of this, it is difficult to link official data about the company with its mentions in the public field and vice versa. There is no reliable tool on the market that finds the brands of a company (and a legal entity by name) with an assessment of reliability. The task is relevant for company analytics, counterparty verification, media monitoring, and any services that work with data on legal entities in Kazakhstan.

A reusable tool will appear that reliably links legal entities (BIN/name) with their brands and trade names. This will improve the quality of data on companies in the Republic of Kazakhstan, allow linking official data with the media field (mass media, social networks, advertising), reduce manual labor in comparing and marking names and reduce the number of errors. The solution is applicable in counterparty verification, media monitoring, and data analytics services. For the IT sector of Kazakhstan, this is a ready-made component of normalization and deduplication of names, which can be reused in different products.

Maminov Ulugbek Vakhidzhanovich