Testing applications with a defined input format
AssigneeGoogle Inc.
InventorsFranjo Ivancic, László Szekeres, Domagoj Babic
Life of this patent
Estimating the term…
Abstract
A system and method are provided for testing the performance of applications. By way of example only, the method may include training a neural network with documents containing text elements that are arranged in accordance with a defined format and using the neural network to determine the predictability of the value of individual text elements within a test document. When the neural network indicates that the value of a text element is unlikely, the value may be modified and the modified document may be used to test an application that processes documents in accordance with the defined format.
Claims18
A method comprising:
- receiving a first sequence of text element values,
- determining, with one or more computing devices, a value of a following text element, wherein determining the value of the following text element comprises:
- determining, based on the first sequence of text element values and a recurrent neural network trained with a corpus of text element values sequenced in accordance with a defined format, a probability value associated with a first text element value, wherein the probability value is related to the predictability of the first text element value following the first sequence of text element values based on a probabilistic model learned by the recurrent neural network from the corpus,
- comparing the probability value to a modification eligible threshold, and
- selecting either the first text element value or a second text element value as the value of the following text element based on whether the probability value exceeds the modification eligible threshold;
- processing, with the one or more computing devices, a second sequence of text element values with a set of instructions after determining that the probability value does not exceed the modification eligible threshold, the second sequence of text element values comprising the first sequence of text element values followed by the selected value of the following text element;
- and
- testing, with the one or more computing devices, a performance characteristic of the set of instructions when the set of instructions processes the second sequence of text element values, wherein testing a performance characteristic comprises determining whether there is a security issue associated with the set of instructions.
A system comprising one or more computing devices, and memory storing instructions executable by the one or more computing devices, wherein the instructions executable by the one or more computing devices comprise:
- receiving a first sequence of text element values,
- determining a value of a following text element, wherein determining the value of the following text element comprises
- determining, based on the first sequence of text element values and a recurrent neural network trained with a corpus of text element values sequenced in accordance with a defined format, a probability value associated with a first text element value, wherein the probability value is related to the predictability of the first text element value following the first sequence of text element values based on a probabilistic model learned by the recurrent neural network from the corpus,
- comparing the probability value to a modification eligible threshold, and
- selecting either the first text element value or a second text element value as the value of the following text element based on whether the probability value exceeds the modification eligible threshold;
- processing a second sequence of text element values with a set of test instructions after determining that the probability value does not exceed the modification eligible threshold, the second sequence of text element values comprising the first sequence of text element values followed by the selected value of the following text element;
- and
- testing a performance characteristic of the set of test instructions when the set of test instructions processes the second sequence of text element values, wherein testing a performance characteristic comprises determining whether there is a security issue associated with the set of instructions.
A non-transitory computing-device readable storage medium on which computing-device readable instructions of a program are stored, the instructions, when executed by one or more computing devices, causing the one or more computing devices to perform a method, the method comprising:
- receiving a first sequence of text element values,
- determining a value of a following text element, wherein determining the value of the following text element comprises
- determining, based on the first sequence of text element values and a recurrent neural network trained with a corpus of text element values sequenced in accordance with a defined format, a probability value associated with a first text element value, wherein the probability value is related to the predictability of the first text element value following the first sequence of text element values based on a probabilistic model learned by the recurrent neural network from the corpus,
- comparing the probability value to a modification eligible threshold, and
- selecting either the first text element value or a second text element value as the value of the following text element based on whether the probability value exceeds the modification eligible threshold;
- processing a second sequence of text element values with a set of test instructions after determining that the probability value does not exceed the modification eligible threshold, the second sequence of text element values comprising the first sequence of text element values followed by the selected value of the following text element;
- and
- testing a performance characteristic of the set of test instructions when the set of test instructions processes the second sequence of text element values, wherein testing a performance characteristic comprises determining whether there is a security issue associated with the set of instructions.
Dependencies are inferred from the claim text.