Software demand error detection network training method, application method and electronic equipment

• Main Authors: LI FUYANG, ZHAO DONGDONG, LI YANHUI, SONG BAILING, YANG PEIXIN, ZHANG ZEXIAN, YU XIAO, HU WENHUA, HE YANZHONG
• Format: Patent
• Language: Chinese; English
• Published: 08.03.2024
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; PHYSICS

The invention discloses a software demand error detection network training method, an application method and electronic equipment. The method comprises the following steps: acquiring a software demand document data set; constructing an initial software demand error detection network, inputting the annotation data, and training to obtain the initial software demand error detection network; selecting a part of unlabeled data, inputting the selected unlabeled data into the preliminary software demand error detection network, predicting to obtain a feature vector set and prediction probability distribution, determining a query strategy value according to the feature vector set and the prediction probability distribution, and selecting to-be-labeled data according to the query strategy value; and marking the to-be-marked data to obtain re-marked data, updating the software requirement document data set, and iteratively training the preliminary software requirement error detection network until the marked resources