International Business Machines Corporation

Peter J Johnson

Fenglian Xu

Timothy John Baldwin

David Locke

Andrew Johnson

George Black

Ben Sasson

Martin J Gale

Barry Neil Dow

Paul G Clarke

Jason C Edmeades

Clare J Owens

William Anthony Kingston

Stephen J Upton

Robin Miles Drew

Roger Meli

For communication pattern recognition, an apparatus is disclosed. The apparatus includes a baseline analysis module that samples predefined sources associated with a user and generates a baseline fingerprint for the user. The apparatus includes an active analysis module that re-samples the predefined sources associated with the user after a predefined time interval and generates an active fingerprint for the user. The apparatus includes a verification module that compares the active fingerprint to the baseline fingerprint and determines whether the active fingerprint closely matches the baseline fingerprint. If the active fingerprint closely matches the baseline fingerprint, then the verification module replaces the baseline fingerprint with the active fingerprint. If the active fingerprint does not match the baseline fingerprint, then the verification module performs a predefined action to rectify differences between the baseline fingerprint and the active fingerprint.

Communication pattern recognition

Method and system are provided for communication monitoring based on sentiment. The method includes: monitoring a new communication prior to sending created by a sender to a recipient, wherein the new communication includes text input by the sender; and analyzing the text of the new communication using sentiment analysis and determining a sentiment score on a scale between negative sentiment and positive sentiment. In response to the sentiment score for the text of the new communication being on the negative side of a predefined threshold on the scale, referencing an overall relationship score based on past communications between the sender and the recipient. In response to the overall relationship score being on the negative side of a predefined threshold, holding a transmission of the new communication for further review.

Communication monitoring based on sentiment

An aspect includes verifying a telephone caller origin carried out at a security application. The verifying includes activating a security process during an active telephone call to a user received at a first call identifier, accessing registered user service providers to obtain contact details of a registered service provider purporting to be the origin of the active telephone call, and sending a query to the obtained contact details of the registered service provider. The query includes the first call identifier. The verifying also includes receiving a returned query result confirming or denying that the telephone call is from the registered service provider.

Verifying telephone caller origin

A method of operating a network of connected nodes comprises the steps of maintaining a trust value for each node in the network, maintaining one or more attributes for each node in the network, each attribute comprising a value and a respective confidence level, and performing a validation of a specific attribute of a specific node. The validation comprises the steps of querying nodes connected to the specific node for specific attribute values for the specific node, creating a results set from the returned attribute values from the connected nodes and the trust values of the connected nodes, and processing the results set according to a predefined algorithm to determine a new value and/or a new confidence level for the specific attribute of the specific node.

Node validation in a network

Embodiments are directed to refining hierarchies in object-oriented models. A method includes providing a business object model in the form of an object-oriented model having one or more members with multiple distinct verbalizations and identifying distinct verbalizations of a given business object model member. The method also includes reviewing existing rules of the business object model to produce mappings of the distinct verbalizations and any attributes or operations used in conjunction with the distinct verbalizations of members of the business object model and analysing the mappings to identify patterns of use of the distinct verbalizations. The method further includes categorising a distinct verbalization as a superclass or subclass.

Refining hierarchies in object-oriented models

The disclosure is for run-time accessing a software component is provided, together with a computer system embodying the same, and a software utility directing the method. Initially a software component is loaded and analyzed to identify one or more features thereof, which features are checked against a database. Following the check, the component is classified as one of at least first, second, and third component types. A first component type is run without amendment. A second component type is replaced by a component of equivalent functionality prior to running. A third component type is blocked from running.

Realtime tracking of software components

A method and system detect a failed communication transmission that fails to reach a recipient after attempting to be transmitted from a sender to the recipient over a communication path. Upon detection of the failed communication, the method initiates a first inquiry process from the sender. This first inquiry process tests how far a first test communication transmission can travel from the sender to the recipient along the communication path. Similarly, upon detection of the failed communication, the method initiates a second inquiry process from the recipient. The second inquiry process tests how far a second test communication transmission can travel from the recipient to the sender along the communication path. The method combines the results of the first inquiry process and the results of the second inquiry process to determine which of the nodes in the communication path are not successfully forwarding test communications to identify at least one possibly faulty node. The identification of the possibly faulty node is then output to the sender and the recipient.

System and method for detecting a network failure

A method of pre-processing a transaction within an application program in a distributed data processing system. The system comprises a network, a first computer system, a second computer system and a communication channel between the computer systems for communicating over the network. Each of the computer systems comprises a processor, memory for storing transactions and an operating system for executing the application program. Firstly, the first computer system sends a transaction, which is received by the second computer system. Next, the second computer system determines whether the transaction is current. If the transaction is current, the second computer system forwards the transaction for processing. If the transaction is not current, the second computer system sends software to the first computer system, in order to update the transaction. Once the first computer system has received the software, it applies the software to the transaction.

Transaction processing in a distributed data processing system

Method of transferring messages between computer programs across a network

Provided are methods and systems for message transfer between nodes of a communication network which reduce the occurrence of repeated transfer failures associated with a particular message or messages within a batch. When a batch of messages sent from a sender node cannot be stored at a receiver node, the receiver node identifies which messages within the batch caused the rejection and communicates this information to the sender node. The sender node records the destinations of the identified messages in a 'destination black list', and then separates messages having these identified problematic destinations from the other messages. The batch of messages is then retransmitted to the receiver, but with the messages which have the identified problematic destinations excluded from the batch. These excluded messages are then dealt with separately without causing repeated rejection of the batch.

Growing community of inventors

Eastleigh, United Kingdom

Peter J Johnson