Abstract:
A solution (A1-A12) is proposed for monitoring usage of bundles (240) in the OSGi environment. For this purpose, an event notification interface (410) of the OSGi environment is exploited. Particularly, each event (A2-A5) relating to a change of state of a bundle (such as its addition, removal, starting and stopping) is detected (A6-A9) by an auxiliary bundle (240a) - which previously subscribed (A1) to the event notification interface. The auxiliary bundle can then forward (A10) the detected events to a licensing agent, which identifies (A11-A12) the bundle by means of a software catalogue (420). In this way, the desired result is achieved without requiring any instrumentation of the bundles to be monitored.
Abstract:
PROBLEM TO BE SOLVED: To reduce quantity of bit map data to be sent by converting a sub-set of a graphic command to a format depending on no operating system and sending it to an address data processing system. SOLUTION: A sever connects a target computer 201 to a control side computer 211. A hook display driver 209 intercepts a graphic command issued from a graphic engine 203 to a display driver 205. The hook display driver 209 converts a graphic command issued from the graphic engine 203 to a format depending on no system using a protocol previously defined. The command depending on no system is sent to the control side computer 211, converted again to a graphic command of a display driver 213 using the same protocol, and reproduced by a display device.
Abstract:
Ein durch einen Computer implementiertes Verfahren zum Trainieren eines Klassifikators von natürlicher Sprache, der einer Chat-Schnittstelle eines Computersystems zugehörig ist, wird bereitgestellt. Zu dem Verfahren kann Empfangen eines Trainingsdatensatzes gehören, der einen ersten Satz von Ausdrücken aufweist, die einer Absicht entsprechen. Zusätzliche Ausdrücke, die der Absicht entsprechen, können erzeugt werden, wobei die zusätzlichen Ausdrücke auf Grundlage des ersten Satzes von Ausdrücken, die der Absicht entsprechen, erzeugt werden. Der Klassifikator von natürlicher Sprache kann auf Grundlage des ersten Satzes von Ausdrücken und der zusätzlichen Ausdrücke, die der Absicht entsprechen, trainiert werden. Der trainierte Klassifikator von natürlicher Sprache kann so implementiert werden, dass er eine Absicht ermittelt, die durch eine erkannte Abfrage ausgedrückt wird, auf Grundlage des ersten Satzes von Ausdrücken und der zusätzlichen Ausdrücke, mit denen der Klassifikator von natürlicher Sprache trainiert wurde.
Abstract:
Enhancing a mind map using a content pool with information items, performing a first semantic scan based on a selected object and generating a new object and a related connection to the selected object based on the semantic scan. A strength value for each of the connections of the first mind map and the connection with the lowest strength value is determined and recalculating the strength value using an (d) external knowledge base to define the connection as obsolete if the strength value decreases. The cycle may be iterated for data from a browser or chat history, a document in a folder or cloud storage, e-mail database, or a content management system. The mind map may then be presented to the user for easy visualisation of links between topics in large data sets, as spider grams.
Abstract:
In a method and system for remote control of a target computer 201 by a controller computer 211 through a network, graphic commands issued by the graphic engine 203 of the target computer are "hooked" 209 and translated into a system independent format. The translated commands are then transmitted to the controller computer 211 to be replayed on the display. The independency from the operating system is realized by defining a protocol which is used to translate a subset of the graphic commands in the target computer before the transmission to the controller computer where the commands are re-translated according to the same protocol.
Abstract:
Eine Lösung zum Erstellen einer Rangfolge einer Mehrzahl von Dokumenten in einem Datenverarbeitungssystem wird vorgeschlagen, wobei jedes der Dokumente eine Mehrzahl von Dokumenttermen aufweist. Ein entsprechendes Verfahren weist Empfangen einer Anfrage auf, die einen oder mehrere Anfrageterme zum Auswählen der Dokumente aufweist, Berechnen entsprechender Ähnlichkeitsindizes zwischen der Anfrage und den Dokumenten, wobei jeder der Ähnlichkeitsindizes auf Grundlage eines Vergleichs zwischen den Anfragetermen und den Dokumenttermen des entsprechenden Dokuments berechnet wird; Bestimmen von entsprechenden Indikatoren für semantischen Reichtum der Dokumente; Anpassen der Ähnlichkeitsindizes auf Grundlage der Indikatoren für semantischen Reichtum, und Erstellen einer Rangfolge von mindestens einem Teil der Dokumente, um diese auf Grundlage der entsprechenden angepassten Ähnlichkeitsindizes auszuwählen.
Abstract:
A method to reduce the number of bits to display a pixel colour, initially represented by a three colour model (e.g. RGB octets). The method comprises reducing the bit pattern of the colour to a lower (e.g. four bit) pattern by (a) analysing the first bit, or first few bits, of each of the three bit patterns, to determine tone, (b) assigning a value to the first bit of the reduced (four) bit pattern, based on the determined tone and (c) assigning the remaining (e.g three) bits of the reduced bit pattern using a predetermined bit (e.g. the second bit) from each of the three original bit patterns. The method has particular application to remote control of computer systems, where a terminal such as 105 is used to control another terminal, 103; the speed/processing complexity of the remote control can be greatly reduced if terminal 105 displays a lower resolution version of that shown on terminal 103.
Abstract:
A method and a system for a fast and reliable color reduction from multiple bit color representations to 4 bit color representations. The method for color reduction in a computer system having a three color model for color representation of a pixel, each color being represented by a multiple bit pattern, reduces the color to a four bit value by analysing the first bit of each of said three multiple bit patterns to determine tone; assigning a value to the first bit of said four bit value, based on said determined tone; and assigning the remaining three bits of said four bit value by utilising one pre-determined bit from each of said three multiple bit patterns.