| CPC G06F 16/45 (2019.01) | 20 Claims |
|
1. A computer-implemented method comprising:
training a neural network model on a creative work database, the creative work database comprising data representative of a plurality of identified creative works and authors associated with the plurality of identified creative works;
further enabling the neural network model, by applying machine learning to the neural network model, with a functionality of identifying a feature within the plurality of identified creative works, the further training producing a trained machine learning model, the feature being at least one of an audio-related feature, a text-related feature, a visual-related feature, and an emotion-related feature;
extracting, using the trained machine learning model, a first set of features from an unidentified creative work;
identifying, using the trained machine learning model, a type of sophistication corresponding to the unidentified creative work, wherein the type of sophistication is based at least in part on a complexity corresponding to the first set of features extracted from the unidentified creative work, and wherein an identification of the type of sophistication narrows down a list of potential authors of the unidentified creative work;
comparing, using the trained machine learning model and within the narrowed down list of potential authors, the first set of features corresponding to the unidentified creative work to a second set of features corresponding to an identified creative work to obtain a similarity score between the unidentified creative work and the identified creative work;
comparing, using the trained machine learning model, the similarity score to a threshold similarity score to determine whether the similarity score meets the threshold similarity score;
upon a determination, by the trained machine learning model, that the similarity score meets the threshold similarity score, identifying a particular author from the narrowed down list of potential authors to associate with the unidentified creative work and classifying the unidentified creative work as a new identified creative work;
authenticating, using the trained machine learning model, the new identified creative work by associating a detail related to the particular author with the new identified creative work; and
updating the creative work database and the trained machine learning model to include the new identified creative work and an authentication reflective of the particular author.
|