Some problems we solve

Click each card to solve the problem.

Automatic extraction of main content

Problem 1 - Precise Search and Selection of Web Pages

A system that searches for information on the Internet based on a query (for example, bank & crisis & mortgage & Ukraine) will find thousands of sources from which to extract information. If we search for these keywords across an entire page, we will obtain many pages where some of these terms are not part of the main content but appear in related news fields, user comments, advertisements, banners, etc. This will lead us to include a large number of irrelevant pages in our search results, resulting in unnecessary processing, wasted time and resources, and, worse yet, incorrect and inaccurate results.

Solution

CESy detects the main content of web pages and ensures that the search terms appear in the primary content of the pages I want to find (and not in other parts of the webpage).

Problem 2 - Relevant Information Extraction

I want to display all today’s news related to a specific company. After conducting an information search across various sources (newspapers, social media, feeds, etc.), a system gathers the most relevant web pages for that search. Unfortunately, these pages often contain irrelevant information (that is not part of the main news content) such as menus, banners, ads, headers, footers, navigation elements, templates, related news, comments, etc.

Solution

CESy extracts only the relevant content from each page, ensuring that we can display or analyze (e.g., sentiment analysis, semantic analysis) only the useful information from each page. Once extracted, an executive who wants to analyze the information can view all the synthesized information at once.

Problem 3 - Information Processing for the Visually Impaired

Visually impaired people face challenges when listening to web content, as screen readers read menus and any other text sequentially on a webpage. This forces the user to listen to a lot of irrelevant information (including ads and inserted promotions) before they can access the main content.

Solution

CESy extracts the primary content before the browser reads the entire webpage, so the browser reads only the desired information. This results in significant time savings for visually impaired users.

Automatic extraction of reviews

Problem 1 - Extraction of Global Reviews

I want to understand my customers’ opinions about my product. However, my product is sold on many different online stores. Therefore, I need to gather reviews (ratings) of my product from various online stores and combine all this information for analysis. Unfortunately, each website displays information differently, and pages frequently change their structure.

Solution

CESy extracts reviews from any page without needing to know its structure. This allows for collecting reviews from all sites and compiling them into a single report.