International Business Machines Corporation

Itai Gordon

Ilan D Prager

Gil Fuchs

Yoav Ben-Yair

Peter Hagelund

Shlomit Avrahami

Oren Paikowsky

Joshua Fox

Kristin E McNeil

Yakir Keisar

Miriam Nizri

Radha Mohan De

Josemina Marcella Magdalen

Oded Sofer

Yifat Yulevich

Alexander Pyasik

David J Henderson

Tali Finelt

Or Bar-Yaacov

Amitai Madar

Radha Radha Mohan De

Elik Katz

Denis Voloshin

Rotem Brosh

Shiran Yosef

Dhruv Khurana

An embodiment for recursively adapting a sensitive content masking technique is provided. The embodiment may include receiving a request from a primary user to share an original document. The embodiment may also include receiving an identity of a secondary user who needs access to a masked version of the original document. The embodiment may further include scanning the original document for sensitive information and identifying sensitive information in the original document. The identified sensitive information may be displayed to the primary user. The embodiment may also include generating a masked value for each piece of identified sensitive information. The embodiment may further include suggesting one or more groups of secondary users if there are additional secondary users. The embodiment may also include presenting the masked version of the original document to the secondary user. The embodiment may further include indexing the masked version of the original document.

Recursively adapting a sensitive content masking technique

In an approach for enhancing an experience of a user listening to and/or watching an audio-visual content by modifying future audio and/or video frames of the audio-visual content, a processor captures a set of sensor data from an IoT device worn by the first user. A processor analyzes the set of sensor data to generate one or more connotations by converting the emotion using an emotional vector analytics technique and a supervised machine learning technique. A processor scores the one or more connotations on a basis of similarity between the emotion exhibited by the first user and an emotion expected to be provoked by a second user. A processor determines whether a score of the one or more connotations exceeds a pre-configured threshold level. Responsive to determining the score does not exceed the pre-configured threshold level, a processor generates a suggestion for the producer of the audio-visual content.

Automatic connotation for audio and visual content using IOT sensors

A method, computer program product and system for generating false data for suspicious users. A suspicious user is identified. Actions of the user are then tracked. The user attempting to access sensitive information is detected. Relevant false sensitive information corresponding to the sensitive information is then detected. The relevant false sensitive information is then mapped to the sensitive information. The relevant false sensitive information is provided to the suspicious user. In response to user input, at least one command is executed, where the at least one command includes the relevant false sensitive information and not the sensitive information.

Generating false data for suspicious users

The capture and search of application properties for applications installed on virtual machines, includes obtaining a plurality of logs on a plurality of virtual machine images coupled to a network. For each given virtual machine image of the plurality of virtual machine images, the method includes: obtaining one or more logs on the given virtual machine image of the plurality of virtual machine images; analyzing each given log of the one or more logs to identify one or more concepts related to a set of application properties for one or more applications installed on the given virtual machine image; associating the one or more concepts and the set of application properties with an identity of the given virtual machine image; and storing the association of the one or more concepts and the set of application properties with the identity of the given virtual machine image in a virtual machine catalogue.

Capture and search of virtual machine application properties using log analysis techniques

A method, computer program product and system for preventing unauthorized access of confidential information. The transmission of data from a first user to a second user is detected. An authorization level corresponding to the second user is then determined. Furthermore, a probability that the authorization level corresponding to the second user and the data is accurate is generated. Additionally, a determination is made that the data includes sensitive information that the second user is not authorized to access based on the authorization level. Moreover, the data can be modified based on the probability, where the data is to be redacted if the probability is within a range of a threshold value or the data is to be blocked from transmission if the probability is above the range.

Monitoring and preventing unauthorized data access

Client side code of a web application can be received by a mobile device. The mobile device can automatically inject mobile aware code into the client side code of the web application. The mobile aware code can modify the client side code of the web application for use by the mobile device, wherein the client side code of the web application is not modified for use by the mobile device prior to the mobile aware code being injected. The client side code of the web application modified by the mobile aware code can be executed by the mobile device.

Adaptation of a web application for use by a mobile device

A computer-implemented method is provided. The method may include determining a behavioral pattern of a user based on historical data access events and historical data access conditions corresponding to the historical data access events, wherein the data access events are associated with a computer enterprise system. A data access request from the user with respect to a secure resource may be received from a computing node connected to the computer enterprise system. A behavioral state of the user may be determined with respect to the data access request and data access conditions corresponding to the data access request. A discrepancy between the behavioral pattern and the behavioral state of the user may be detected. A security risk level may be determined based on the discrepancy. In response to determining that the security risk level exceeds a predetermined threshold, a security action may be performed with respect to the secure resource.

Emotion-based database security

Detecting malware attacks is described herein. A computer-implemented method may include receiving, via a processor, events from a plurality of activity monitors. The method also include extracting, via the processor, a plurality of behavioral features from the received events. The method may further include detecting, via the processor, a malware attack based on the extracted behavioral features using a malware identification model trained on private data and public data using a machine learning technique, wherein the private data includes private enterprise attack findings. The method may also include executing, via the processor, an ad hoc protection improvement based on the detected malware attack.

Detecting malware attacks using extracted behavioral features

Data lineage including a plurality of levels can be received. A configuration also can be received. A three dimensional (3D) virtual reality (VR) model can be built, the 3D VR model including a plurality of floors based on data lineage content corresponding to the plurality of levels and the configuration. The 3D VR model can depict, on at least a first of the plurality of floors, a plurality of rooms of a virtual building representing data elements and hallways of the building representing data flows between data elements. A view of the 3D VR model can be displayed on a display device, wherein the 3D VR model is configured for a user to navigate the plurality of the rooms and hallways of the virtual building to determine lineage of data.

Displaying data lineage using three dimensional virtual reality model

Receiving contextual data including a facial movement associated with an active document. A response associated with the active document is detected and associated with the received contextual data. A contextual metadata tag is generated based on the detected response to the active document. A contextual keyword is created that corresponds to the contextual metadata tag. Search results received in response to the query are filtered based on the contextual metadata tag.

Filtering document search results using contextual metadata

Embodiments of the present invention disclose methods and systems which receive a user credential corresponding to a user, a task to be performed by the user, a security policy including a user role, and sensitive information. These methods and systems dynamically provision virtual machines including un-redacted information from received sensitive information. Furthermore, a set of tools process the redacted information, based on the user credential, the task to be performed, and the security policy.

Dynamically provisioning virtual machines

Disclosed embodiments provide techniques for accessing a document from a cloud storage system and controlling the display of sensitive data within the document based on user permissions. One or more restricted information segments are identified within a document to be stored on the cloud storage system. Restricted information segments can include anything within an electronic file for which it is desired to provide multiple levels of access. In some embodiments, the restricted information segments are automatically identified via computer-implemented natural language processing (NLP) techniques. For each restricted information segment, one or more alternative data sequences are generated. The alternative data sequences are encrypted using various keys residing on a client device associated with a user. The keys can be used to decrypt data stored within a multiple-value encrypted field structure. In this way, multiple levels of information can be disseminated to an audience with various permission/privilege levels.

Remote file storage with multiple access levels

A developer is monitored by at least one sensor, and developer data is gathered from the at least one sensor. Code change data is gathered for changes made by the monitored developer to at least one area of code, and the code change data is mapped to the developer data. Test cases are run to test the at least one area of code, and to identify failed test cases. Code change data corresponding to the failed test cases is also identified, as well as developer data mapped to the corresponding code change data. Further, a prediction model that correlates test case failure with the developer data is generated. The prediction model is used to generate probabilities of failure for test cases based on a new developer data mapped to new code test data.

Growing community of inventors

Jerusalem, Israel

Itai Gordon