Abstract: Systems and methods may reduce bias in an artificial intelligence model. The system may receive word embedding model generated based on a corpus of words. The system may determine a bias definition vector in an embedding space of the word embedding model. The system may receive bias classification criteria. The bias classification criteria may include logic to group word vectors in the word embedding model based on a distance measurement from the bias definition vector. The system may identify, in the word embedding model, a first group of vectors and a second group of vectors based on the bias classification criteria and the bias definition vector. The system may generate a debiased artificial intelligence model. The debiased artificial intelligence model may include associations between words and metrics. The system may weight the metrics for the words associated with the first and second group of vectors with a non-zero penalization factor.

Abstract: Implementations include providing a first set of tags by processing a document using generic entity extraction based on one or more external taxonomies, providing a second set of tags by processing the electronic document using specific entity extraction based on internal taxonomies specific to the enterprise, determining a relevance score for each tag in the first set of tags, and the second set of tags, defining a set of tags including one or more tags of the first set of tags, and one or more tags of the second set of tags, tags of the set of tags being in rank order based on respective relevance scores, receiving user input to the set of tags, and performing one or more of adjusting a ranking of tags based on the user input, and editing at least one internal taxonomy of the one or more internal taxonomies based on the user feedback.

Abstract: Systems and methods may reduce bias in an artificial intelligence model. The system may receive word embedding model generated based on a corpus of words. The system may determine a bias definition vector in an embedding space of the word embedding model. The system may receive bias classification criteria. The bias classification criteria may include logic to group word vectors in the word embedding model based on a distance measurement from the bias definition vector. The system may identify, in the word embedding model, a first group of vectors and a second group of vectors based on the bias classification criteria and the bias definition vector. The system may generate a debiased artificial intelligence model. The debiased artificial intelligence model may include associations between words and metrics. The system may weight the metrics for the words associated with the first and second group of vectors with a non-zero penalization factor.

Abstract: The systems and methods to reduce bias in an artificial intelligence model are provided. The system may receive word embedding model generated based on a corpus of words. The system may determine a bias definition vector in an embedding space of the word embedding model. The system may receive bias classification criteria. The bias classification criteria may include logic to group word vectors in the word embedding model based on a distance measurement from the bias definition vector. The system may identify, in the word embedding model, a first group of vectors and a second group of vectors based on the bias classification criteria and the bias definition vector. The system may generate a debiased artificial intelligence model. The debiased artificial intelligence model may include associations between words and metrics. The system may weight the metrics for the words associated with the first group of vectors and second group of vectors with a non-zero penalization factor.

Abstract: Implementations include providing a first set of tags by processing a document using generic entity extraction based on one or more external taxonomies, providing a second set of tags by processing the electronic document using specific entity extraction based on internal taxonomies specific to the enterprise, determining a relevance score for each tag in the first set of tags, and the second set of tags, defining a set of tags including one or more tags of the first set of tags, and one or more tags of the second set of tags, tags of the set of tags being in rank order based on respective relevance scores, receiving user input to the set of tags, and performing one or more of adjusting a ranking of tags based on the user input, and editing at least one internal taxonomy of the one or more internal taxonomies based on the user feedback.