Advanced online news analytics and verification endpoints
The TrustServista REST API allows developers to fully leverage our state-of-the-art algorithms for Deep Content analytics and verification purposes.
Text Analysis API endpoints
Extract clean text from an URL or raw HTML
. Process any HTML page and obtain only the useful content from it by removing unwanted page regions or ads.
Automatically generate a summary for an URL or raw text
. Understand the meaning of web content by reading only the key sentences of the whole page text.
Extract named entities from an URL or raw text
. Named entities, categorized by type (people, organization, locations, products, other), constitute the core factual information of any content.
Calculate the semantic similarity between two URLs or pieces of raw text
. The semantic similarity finds identical meaning contained by the analyzed content pieces, ignoring syntax or grammar.
Extract the sentiment from an URL or raw text
. Determine the polarity of an article: negative, neutral, positive.
Classify the content from an URL or raw text
. Automated classification using standard IPTC and IAB taxonomies.
Detect the language of a given URL or raw text. Support for over 70 languages, including all Latin-based and most Asian languages.
Vectorization of a given URL or raw text . Returns a vector of floating point numbers corresponding to the content.
Verification and metadata extraction for an URL or raw text.
Retrieve key information relevant to trustworthiness, such as content author, quality score, named entities, Patient Zero and more.
News Analytics API endpoints
Return an article’s relationships to other articles
. The relationship graph will contain up to the 3rd degree implicit, explicit and semantic links.
Generate an article-centered story for a given time interval
. Understand how many similar articles were published in the last day or week.
Generate topics (stories) from articles processed in a given time interval
. Clustering articles based on their similarity for the last hour or day.
Historical search in our news database
. Retrieve articles of interest by using a complex and versatile search query.
Relevant statistics for any website processed by TrustServista
. Content Quality and Clickbait metrics averages, sentiment distribution, volumes of published articles, for a given time interval.
Generate the content quality score for an URL or raw text
. The aggregate score takes in account multiple metrics relevant to the content trustworthiness.
Determine the original source of information for a URL or raw text: Patient Zero
. P0 is found using complex graphs built with link analysis and semantic similarity algorithms.