| CPC G06F 11/3608 (2013.01) [G06F 8/10 (2013.01)] | 20 Claims |
|
1. A computer-based method of proactively identifying potential errors in code instance data for creating technical documentation comprising:
receiving code instance data;
generating potential error classifications for the received code instance data using a convolutional neural network and natural language processing techniques, wherein using the convolution neural network and the natural language processing techniques further comprises applying algorithms including pattern matching to identify patterns in specific keywords and text, image and diagram processing to identify the potential errors in images and diagrams, and word structure processing to determine word meaning and use in phrases and sentences to further identify the potential errors;
training one or more machine learning (ML) algorithms to perform a correlation analysis to derive correlations between the generated potential error classifications for the received code instance data and similarly occurring classifications in one or more historical code instances;
calculating a score for each of the derived correlations from the trained one or more ML algorithms, the calculated score corresponding to a likelihood that the received code instance data is correlated to the similarly occurring classifications and represents a potential error; and
outputting notifications to a user for each of the derived correlations based on a predefined threshold value for the calculated score, further comprising outputting a notification for a derived correlation based on the calculated score for the derived correlation exceeding the predefined threshold value, and wherein the outputting further comprises generating and providing a link to the code instance data for the derived correlation.
|