🔢Data Input
Last updated
Last updated
By Data Input we mean the processing of all user submissions (news or social posts) incoming to Confidently for validation.
The bulk of the content shown on the platform will be user submissions, which become validation requests and then converted into a pool of work.
The resource gets analyzed to find duplicate submissions or if there's any similar or identical resource posted
A snapshot of the web page or text contents is performed and timestamped using external scraping services
Category is added to the resource either automatically (via text analysis) or suggested by the submitting user. Categorization is key to find suitable validators and supervalidators
Expert Engine is invoked to perform a pre-scoring and discard the extremes of the spectrum, that is news that are blatantly fake or satirical and don't need further community discussion
The resource enters the validation feed and is presentend in frontend to validators and supervalidators
The vast majority of misleading and fake news circulate through social media or via websites that the Iffy Index defines as low/very low factual, problematic, unverified and similar. Our focus as Confidently project is exactly on these kind of news, since they offer an ample margin for debate due to their clickbait or politically biased nature.
If the same URL is submitted twice, the user is sent to the already open pool. If a news is relayed by multiple newspapers, we suggest the user to post the main source of content (e.g. press agency) or the oldest available. We're working on finding a way to suggest similar already posted news starting from a news URL. These steps are really key to prevent "dispersing" validator and supervalidator efforts across pools that are essentially equivalent in contents.
Resources (i.e. news, social posts) to be validated are presented to the users using a matchmaking criteria based on the news category and the user expertise area which they selected during signup or profile edit. To make navigation easier, users will be able to explore pools containing new and trendin topics, that is topics that occur with a high frequency among the recently submitted resources.
