Abstract: Conventional Question and Answer (QA) datasets are created for generating factoid questions only and the present disclosure generates longform technical QA dataset from textbooks. Initially, the system receives a technical textbook document and extracts a plurality of contexts. Further, a first plurality of questions are generated based on the plurality of contexts. A plurality of answerable questions are generated further based on the plurality of contexts using an unsupervised template-based matching technique. Further, a combined plurality of questions are generated by combining the first plurality of questions and the plurality of answerable questions. Further, an answer for the combined plurality of questions are generated using an autoregressive language model and a mapping score is computed. Further, a plurality of optimal answers are selected based on the corresponding mapping score.

Abstract: A method for predicting software failure characteristics is discussed. The method includes accessing failure data and context data related to a failed software call into a software stack. The failure data indicates software call information. The context data indicates characteristics including functionality of the failed software call. The method includes accessing cluster data on previous software failures. The cluster data includes analyzed failure data and analyzed context data on software call traces. The analyzed failure data is provided by first and second failure tools. The cluster data for software calls is generated through respective instances of the software stack for each software call trace. The method also includes correlating the failed software call trace to a particular software call trace of the cluster data. The correlating is based at least on analysis of clusters indicated by the cluster data, the failure data, and the context data.

Abstract: Various systems, mediums, and methods may involve data engines configured to generate results associated with one or more entities based on neural network configurations. An exemplary system includes a data engine with a training module, a working module, an incremental training module, and a neural network. The data engine may process data associated with the one or more entities and transfer the processed data to an input layer of the neural network. Further, outputs from the input layer may be transferred to a hidden layer of the neural network. Yet further, outputs from the hidden layer may be transferred to an output layer of the neural network. As such, one or more results may be generated from an output layer of the neural network. The one or more results may include an assessment score of the one or more entities.

Abstract: A method for predicting software failure characteristics is discussed. The method includes accessing failure data and context data related to a failed software call into a software stack. The failure data indicates software call information. The context data indicates characteristics including functionality of the failed software call. The method includes accessing cluster data on previous software failures. The cluster data includes analyzed failure data and analyzed context data on software call traces. The analyzed failure data is provided by first and second failure tools. The cluster data for software calls is generated through respective instances of the software stack for each software call trace. The method also includes correlating the failed software call trace to a particular software call trace of the cluster data. The correlating is based at least on analysis of clusters indicated by the cluster data, the failure data, and the context data.

Abstract: Various systems, mediums, and methods may involve data engines configured to generate results associated with one or more entities based on neural network configurations. An exemplary system includes a data engine with a training module, a working module, an incremental training module, and a neural network. The data engine may process data associated with the one or more entities and transfer the processed data to an input layer of the neural network. Further, outputs from the input layer may be transferred to a hidden layer of the neural network. Yet further, outputs from the hidden layer may be transferred to an output layer of the neural network. As such, one or more results may be generated from an output layer of the neural network. The one or more results may include an assessment score of the one or more entities.