Digital Financial Reporting

A recommender system or a recommendation system is an automated system that makes recommendations based on metadata and other information the recommender system has access to.

Recommender systems can make use of collaborative filtering, content-based filtering, and/or metadata stored in a knowledge graph. 

• The collaborative filtering approach build a model from a user's past behavior as well as similar decisions made by other users. This model is then used to predict items that the user may have an interest in.
• A content-based filtering approach uses application state information to help figure out what the application user is trying to do.
• Metadata stored in a knowledge base can be used by a recommender system to help a software application make decisions.

Recommender systems can combine these approaches to create a hybrid system.

How could a recommender system be used in XBRL-based digital financial reporting? Here are some examples: 

• Help an accountant select a financial reporting scheme.
• Help an accountant pick a reporting style from a financial reporting scheme.
• Help an accountant pick a disclosure they want to construct for a financial reporting scheme using some specific reporting style.
• Help an accountant construct a structure or type of structure.
• Help an accountant construct a rule.
• Help an accountant select XBRL taxonomy report elements that would be used to represent a specific disclosure per their report model.
• Help an accountant pick between reporting disclosure alternatives.
• Help an accountant understand which disclosures are required to be provided in their report.
• Report creation wizard.
• Report disclosure wizard.
• Report agenda.
• Report information search.
• Template search.
• Exemplar search.

Recommender systems can use a combination of rule-based information (deductive reasoning) or patterns-based information (inductive reasoning or machine learning) to make recommendations.

Here are three videos that help you understand more about recommender systems:

• Introduction to Recommender Systems
• How Recommender Systems Work
• How does Netflix recommend movies? Matrix Factorization
• How to Design and Build a Recommendation System Pipeline in Python

Recommender systems work using semantic information rather than basic search of text. A recommender system is a type of expert system functionality.

##########################

The Emancipation Of Expert Knowledge