NotAStartupAnymore
Detection of new large companies, super growth and company relocation through webscrapping and language modelling.
Although many companies publish it in the non-financial part of their annual reports, it is not easy to know the exact number of employees of a company, if it has experienced rapid growth or if it has decided to change its registered office and/or head office.
For strategic reasons (sales, operations, etc.) you would like to be able to obtain a report, for a certain period of time, on newly created companies with more than a certain number of employees (suggested, 400), that have experienced a growth also of a specific number (e.g. 300) or also that have changed their headquarters or head office location (city, country, etc.).
It is therefore desired to build a system that, based on information obtained from social networks and news sites, is able to detect these three types of events. To this end, the project will consist of three different modules:
- An aggregation/webscrapping module that will be responsible for collecting the information necessary for the operation of the system. This module will be configurable with different sources and open to future expansion.
- An artificial intelligence system, based on language models (LLMs, SLMs) and/or field segmenters (RPA) to extract the relevant information.
- A report and alert manager that will alert when the indicated events occur and will also issue a report (PDF) on all the companies that have undergone any of these changes in the desired time period.
The system should be able to operate in an unattended/automatic way, periodically scanning the sources of information to be taken into consideration and issuing the corresponding reports and alerts. It is suggested to additionally develop a small web application (backend/frontend) to be able to manage the configuration of the system (limit values of employees, time periods, subscriptions, etc).
- Complexity:8/10
- Technical specifications:Beautiful Soup, Requests
- CATEGORY 2024-2025 , Salamanca Univ.
- TAGS Python , React , Scrapy





