Abstract: Example embodiments are disclosed of systems and methods for predicting failure probabilities of future product tests of a testing sequence based on outcomes of prior tests. Predictions are made by a machine-learning-based model (MLM) trained with a set of test-result sequence records (TRSRs) including test values and pass/fail indicators (PRIs) of completed tests. Within training epochs over the set, iterations are carried out over each TRSR. Each iteration involves sub-iterations carried out successively over test results of the TRSR. Each sub-iteration involves (i) inputting to the MLM values of a given test and those of tests earlier in the sequence while masking those later in the sequence, (ii) computing probabilities of test failures for the masked tests found later in the sequence than the given test, and (iii) applying the PFIs of test results later in the sequence than the given test as ground-truths to update parameters of the MLM.

Abstract: Example embodiments are disclosed of systems and methods for predicting failure probabilities of future product tests of a testing sequence based on outcomes of prior tests. Predictions are made by a machine-learning-based model (MLM) trained with a set of test-result sequence records (TRSRs) including test values and pass/fail indicators (PRIs) of completed tests. Within training epochs over the set, iterations are carried out over each TRSR. Each iteration involves sub-iterations carried out successively over test results of the TRSR. Each sub-iteration involves (i) inputting to the MLM values of a given test and those of tests earlier in the sequence while masking those later in the sequence, (ii) computing probabilities of test failures for the masked tests found later in the sequence than the given test, and (iii) applying the PFIs of test results later in the sequence than the given test as ground-truths to update parameters of the MLM.

Abstract: An embodiment involves obtaining a tabular data set with columns that characterize items relating to behavior of components of a communication network; constructing a frequent-pattern tree, each node being associated with: (i) an item-name for representing an item, (ii) a count of transactions from a root node of the tree to the respective node, and (iii) node-links that refer to other nodes in the tree that represent items having the same item-name; traversing the tree to identify a set of nodes with counts greater than a predefined support threshold; generating, from the nodes, association-rules that are based on antecedent items associated with a target item; reducing the association-rules by (i) removing the association-rules in which the antecedent items thereof are a superset or subset of the antecedent items of a further association-rule, or (ii) combining the association-rules that have antecedent items that are at least partially disjoint and conditionally dependent.

Abstract: An embodiment may involve obtaining a set of data records including features characterizing operational aspects of a communication network. Each data record may include a feature vector and performance metrics of the communication network. Each feature vector may include a multiple elements corresponding to feature-value pairs. A first statistical analysis may be applied to the set of data records and their performance metrics to identify major contributors to degraded network performance. A second statistical analysis may be applied to identify elements that negatively influence the major contributors, and to discriminate between additive effects and incompatibilities as the source of negative influence. For each major contributor, a hierarchical dependency tree may be constructed with the major contributor as the root node and influencer elements as other nodes.

Abstract: An embodiment may involve obtaining a set of data records including features characterizing operational aspects of a communication network. Each data record may include a feature vector and performance metrics of the communication network. Each feature vector may include a multiple elements corresponding to feature-value pairs. A first statistical analysis may be applied to the set of data records and their performance metrics to identify major contributors to degraded network performance. A second statistical analysis may be applied to identify elements that negatively influence the major contributors, and to discriminate between additive effects and incompatibilities as the source of negative influence. For each major contributor, a hierarchical dependency tree may be constructed with the major contributor as the root node and influencer elements as other nodes.