International Business Machines Corporation

Geetha Adinarayan

Gandhi Sivakumar

Shaw-Ben Shi

Meng Hong Tsai

Sattwati Kundu

Mansoor Ahmed

Akshat Dixit

Nair Raghunath Eledath

Dinesh Radhakrishnan

Rohit Pradeep Shetty

Diptikalyan Saha

Anindya Neogi

Vitobha Munigala

Vinay G Rajagopal

Faried Abrahams

Raman Harishankar

Joern Ploennigs

Sathiskumar Palaniappan

Andrew John Bernoth

Amit Mohan Mangalvedkar

Suranjana Samanta

Hari Hara Prasad Viswanathan

Sateesh S Kannegala

Abhishek Mitra

A system, computer program product, and method are presented for providing replacement data for data in a time series data stream that has issues indicative of errors, where the data issues and the replacement data are related to one or more KPIs. The method includes determining one or more predicted replacement values for potentially erroneous data instances in the time series data stream. The method further includes resolving the potentially erroneous data instances with one predicted replacement value of the one or more predicted replacement values in the time series data stream.

Data quality-based computations for KPIs derived from time-series data

Data from a knowledge graph associated with an enterprise may be obtained. The knowledge graph may include a plurality of node descriptors that indicate locations, assets, and sensor data feeds of the enterprise, and a plurality of relationship descriptors that indicate relationships amongst the locations, the assets, and the sensor data feeds of the enterprise. An anomaly detection or prediction model associated with a selected one of the plurality of node descriptors may be auto-created based on the data from the knowledge graph. In the auto-creation, one of a plurality of model types for the anomaly detection or prediction model may be selected based on an identified node type of the selected node descriptor.

Anomaly detection of physical assets by auto-creating anomaly detection or prediction models based on data from a knowledge graph of an enterprise

Methods, systems and computer readable media are provided for configuring machine learning systems to automatically and dynamically select a machine learning model based on statistical profiling of received data to improve machine learning applications for high variance data. Data is received from a system in operation. A profile is computed for the received data. A database comprising a plurality of stored profiles for a dataset is accessed, wherein each stored profile corresponds to a distinct pattern identified in the dataset. The computed profile is compared to each of the stored profiles to determine whether the computed profile matches one or more of the stored profiles. When one or more stored profiles match the computed profile, a matching profile is selected by the machine learning system. The received data is processed using a ML model associated with the matching profile.

Dynamically applying machine learning models from profiling received data

A system, computer program product, and method are presented for providing confidence values for replacement data for data that has issues indicative of errors, where the data issues, the replacement data, and confidence values are related to one or more KPIs. The method includes identifying one or more potentially erroneous data instances and determining one or more predicted replacement values for the potentially erroneous data instances. The method further includes determining a confidence value for each predicted replacement value and resolving the one or more potentially erroneous data instances with one predicted replacement value of the one or more predicted replacement values. The method also includes generating an explanatory basis for the resolution of the one or more potentially erroneous data instances.

Data quality-based confidence computations for KPIs derived from time-series data

A computer-implemented method, system and computer program product for identifying anomalies in data during a data outage. An anomaly detection model is built using data received from a sensor at a characterized granularity. Once a period of service occurs following a data outage, a quantum of missing data during the data outage is identified using predictive modeling if the data during the data outage is not available at the granularity in which the anomaly detection model is built. The identified quantum of missing data is retrofitted into a predicted pattern during the data outage and the analytics are then re-run on the retrofitted quantum of missing data in the predicted pattern to identify anomalies during the data outage. In this manner, anomalies in data, such as data from sensor readings, can be identified during the data outage thereby enabling the model to provide more accurate predictions of anomalies occurring during the data outage.

Identifying anomalies in data during data outage

Aspects of the present invention disclose a method for generating a reasoning query based on a user selection of a generated data visualization of a knowledge graph. The method includes one or more processors generating a knowledge graph of a domain. The method further includes constructing a hierarchy of the knowledge graph. The method further includes generating a data visualization of the domain based at least in part on the hierarchy of the knowledge graph. The method further includes identifying a user selection of one or more nodes of the data visualization. The method further includes generating a reasoning query corresponding to the domain based on the data visualization of the domain and the user selection. The method further includes determining whether the knowledge graph includes a collection of nodes that are on a level of the constructed hierarchy that corresponds to a level of the one or more nodes.

Auto generating reasoning query on a knowledge graph

Provided are techniques for building and fixing a dynamic application topology. Log files are received from multiple sources comprising any of services and nodes. Information is extracted from the log files. An application topology is created for a particular point in time for an application that provides hierarchical relationships of components for executing the application using the extracted information. One or more problems in the application topology are identified. A solution is applied to each of the one or more problems.

Building and fixing a dynamic application topology in a cloud based environment leveraging log file data

In an automatic context adaptive search and result generation, a server obtains an email corpus of a user for a current time period. The server identifies a set of triggering semantics in the email corpus, and using an ontology, identifies a set of topic-context pairs corresponding to each triggering semantic. The server identifies a set of paths in the ontology activated by the set of topic-context pairs and compares the set of activated paths with paths in each heatmap of a set of heatmaps, where each heatmap corresponds to a document in a set of documents. The server identifies one or more heatmaps of the set of heatmaps including one or more paths matching an activated path of the set of activated paths. The server then outputs a search result including the one or more documents corresponding to the one or more heatmaps.

Automatic context adaptive enterprise search and result generation on behalf of a user

A method and system for controlling unwanted phone calls. In response to a determination that a phone number of a current incoming call to a user is not a phone number in a contact list including phone numbers of the user's contacts and to a determination that the phone number of the current incoming call is a phone number in a phone list of a shared table, the shared table for the previous call duration is analyzed for the previous call duration and the previous sentiment of the user during a previous incoming phone call for the phone number of the current incoming call to the user. It is ascertained, from analyzing the shared table for the previous call duration, that the previous call duration is less than a predetermined call duration and the previous sentiment is a negative sentiment, and in response the current incoming call is rejected.

Identifying and controlling unwanted calls

One embodiment for determining a marketing incentive for a user of an electronic device. In one embodiment, a computer processor detects a first electronic device within a retail environment utilizing a second electronic device that also identifies information associated with the first electronic device. In one embodiment, a computer processor determines a behavior associated with the first electronic device based, at least in part, on movement of the first electronic device within the retail environment. In one embodiment, a computer processor identifies data associated with the retail environment that includes information associated with a retailer associated with the retail environment and information associated with the first electronic device. In one embodiment, a computer processor determines a first marketing incentive based, at least in part, on the determined behavior associated with the first electronic device and the identified data associated with the retail environment.

Personalized marketing incentives based on historical information and mobility monitoring

The method include receiving, by one or more computer processors, a request for data usage, the request for data usage being associated with a mobile device. The method further includes identifying, by one or more computer processors, a current data usage quantity, the current data usage quantity being associated with the mobile device. The method further includes determining, by one or more computer processors, whether the current data usage quantity is beyond a threshold data usage. The method further includes responsive to the data usage quantity being beyond the threshold data usage, receiving, by one or more computer processors, mobile device information associated with the mobile device. The method further includes analyzing, by one or more computer processors, the mobile device information. The method further includes determining, by one or more computer processors, access for the request based on the mobile device information and the current data usage quantity.

Data usage recommendation generator

A method, system and computer program product for analytically selecting which tests are to be executed in a continuous delivery process. An analytics processing system constructs a code dependency tree to analyze the tests that are affected by changes in code after a new build is generated. After analyzing the code dependency tree, the system eliminates those tests in the code dependency tree that do not depend on the changed code. The system then analyzes the historical execution records for those tests that have not been eliminated for execution to obtain an execution rate and a failure rate for those tests. A score is generated for each of the tests from the code dependency tree that were not eliminated for execution based on the historical execution rate and failure rate of the test. Tests that have a score that exceeds a threshold are included for execution.

Analytically selecting which tests are to be executed in a continuous delivery process

Systems, methods and tools for leveraging computer networks, data sharing, data analytics and sentiment analysis to control the receipt of unwanted telephone calls and preventing unwanted telephone calls from disturbing a user through the implementation of contacts list and/or a shared table of unique calls. Contact lists and shared tables of unique calls may identify incoming calls that may be directed to the user or filtered out based on whether the incoming call's number is in the user's contact list and/or based on the past experiences of the individual users engaging with the incoming caller. Embodiments may track the past call durations for unique incoming number and track the sentiment of users during these calls and use the collected call data and sentiment determine whether subsequent incoming calls from a phone number should be forwarded to the user's telephone enabled device or automatically responded to with further instructions.

Growing community of inventors

Bengaluru, India

Geetha Adinarayan