Mobile phone with system failure prediction using long short-term memory neural networks
Mobile phones and methods for mobile phone failure prediction include receiving respective log files from one or more mobile phone components, including at least one user application. The log files have heterogeneous formats. A likelihood of failure of one or more mobile phone components is determined based on the received log files by clustering the plurality of log files according to structural log patterns and determining feature representations of the log files based on the log clusters. A user is alerted to a potential failure if the likelihood of component failure exceeds a first threshold. An automatic system control action is performed if the likelihood of component failure exceeds a second threshold.
1. A method for mobile phone failure prediction, comprising:
receiving respective log files from one or more mobile phone components, including at least one user application, wherein the log files have heterogeneous formats;
training a long short-term memory recurrent neural network from training data and user labels on failure time periods;
determining a likelihood of a failure of one or more mobile phone components based on the received log files by clustering the log files according to structural log patterns to form log clusters and determining feature representations of the log files based on the log clusters by processing the feature representations using the long short-term memory recurrent neural network;
alerting a user to a potential failure if the likelihood of the failure exceeds a first threshold; and
performing an automatic system control action if the likelihood of the failure exceeds a second threshold.
2. The method of claim 1 , wherein determining the feature representations comprises determining two, and only two, feature representations.
3. The method of claim 2 , wherein the two feature representations consist of pattern distribution among clustered logs and term frequency inverse document frequency.
4. The method of claim 1 , wherein clustering the plurality of log files comprises performing unsupervised, hierarchical clustering.
5. The method of claim 1 , wherein determining the likelihood of the failure of one or more of the mobile phone components comprises determining a binomial distribution based on an output of the neural network and a binary target vector with two complementary classes.
6. The method of claim 1 , wherein the heterogeneous log formats comprise different delimiter formats.
7. A method for mobile phone failure prediction, comprising:
clustering a plurality of log files from respective mobile phone components, including at least one user application, having heterogeneous log formats according to structural log patterns using unsupervised, hierarchical clustering to form log clusters;
determining feature representations of the log files based on the log clusters, wherein the feature representations consist of pattern distribution among clustered logs and term frequency—inverse document frequency;
training a long short-term memory recurrent neural network from training data and user labels on failure time periods;
determining a likelihood of a failure of one or more of the mobile phone components based on the feature representations using the long short-term memory neural network, a binomial distribution based on an output of the long short-term memory neural network, and a binary target vector with two complementary classes by processing the feature representations using the long short-term memory recurrent neural network;
alerting a user to a potential failure if the likelihood of the failure exceeds a first threshold; and
performing an automatic system control action if the likelihood of the failure exceeds a second threshold.
8. A mobile phone, comprising:
a processor;
a memory configured to store respective log files from one or more mobile phone components, including at least one user application, wherein the log files have heterogeneous formats;
a long-short term memory recurrent neural network that is trained from training data and user labels on failure time periods;
a pattern learning module configured to use the processor to cluster the log files according to structural log patterns to form log clusters;
a feature extraction module configured to determine feature representations of the log files based on the log clusters;
a failure prediction module configured to determine a likelihood of a failure of one or more of the mobile phone components based on the feature representations by processing the feature representations using the long short-term memory recurrent neural network;
an alert module configured to alert a user to a potential failure if the likelihood of the failure exceeds a first threshold; and
a system control module configured to perform an automatic system control action if the likelihood of the failure exceeds a second threshold.
9. The system of claim 8 , wherein the feature extraction module is further configured to determine two, and only two, feature representations.
10. The system of claim 9 , wherein the two feature representations consist of pattern distribution among clustered logs and term frequency inverse document frequency.
11. The system of claim 8 , wherein the pattern learning module is further configured to perform unsupervised, hierarchical clustering.
12. The system of claim 8 , wherein the failure prediction module is further configured to determine a binomial distribution based on an output of the neural network and a binary target vector with two complementary classes.
13. The system of claim 8 , wherein the log files comprise a plurality of different log formats.
14. The system of claim 13 , wherein the heterogeneous log formats comprise different delimiter formats.