Computer-implemented method for processing text input from user

Abstract

According to an embodiment, a computer-implemented method for processing a text input from a user com- prises: obtaining a text input from a user; determining a context of the text input; providing the text input and the context of the text input to a machine learning model; obtaining a first response from the machine learning model; and providing a second response to the user based at least on the first response.

Description

COMPUTER- IMPLEMENTED METHOD FOR PROCESSING TEXT INPUTFROM USERTECHNICAL FIELD

[0001] The present disclosure relates to processing of a text input from a user, and more particularly to a method for proces sing a text input from a user, a computing device , and a computer program product .BACKGROUND

[0002] An increasing number of machine learning (ML) services are being provided for various applications . However, usually these ML models , such as large language models (LLMs ) , have been trained using generic data . Thus , in many applications , the output provided by these generic ML models can be unsatisfactory .SUMMARY

[0003] This summary is provided to introduce a selection of concepts in a s implif ied form that are further described below in the detailed description . This summary is not intended to identify key features or essential features of the claimed subj ect matter, nor is it intended to be used to limit the scope of the claimed subj ect matter .

[0004] It is an obj ective to provide a method for processing a text input from a user , a computing device , and a computer program product . The foregoing and otherobj ectives are achieved by the features of the independent claims . Further implementation forms are apparent from the dependent claims , the description, and the figures .

[0005] According to a first aspect, a computer-implemented method for processing a text input from a user comprises : obtaining a text input from a user ; determining a context of the text input ; providing the text input and the context of the text input to a machine learning model ; obtaining a first response from the machine learning model ; and providing a second response to the user based at least on the first response . The method can, for example , improve responses provided by the machine learning model .

[0006] In an implementation form of the first aspect , the determining the context of the text input comprises : obtaining metadata about the text input ; and determining the context of the text input based at least on the metadata about the text input . The method can, for example , efficiently determine the context of the text input based at least on the metadata about the text input .

[0007] In another implementation form of the first aspect , the metadata comprises static metadata . The method can, for example , efficiently determine the context of the text input based on the static metadata .

[0008] In another implementation form of the first aspect , the static metadata comprises at least one of : information about a service the user is providing thetext input for ; and / or information about a service the user is using to provide the text input . The method can, for example , efficiently determine the context of the text input based on the service the user is using .

[0009] In another implementation form of the first aspect , the metadata comprises dynamic metadata . The method can, for example, efficiently determine the context of the text input based on the dynamic metadata .

[0010] In another implementation form of the first aspect , the dynamic metadata comprises at least one of : time of day of obtaining the text input ; date of obtaining the text input : day of week of obtaining the text input ; time of year of obtaining the text input ; weather information ; status and / or error information ; and / or metadata relating to the user . The method can, for example , efficiently determine the context of the text input .

[0011] In another implementation form of the first aspect , the providing the second response to the user based at least on the first response comprises : modifying the first response ; and providing at least the modified first response to the user in the second response . The method can, for example , improve the response provided by the machine learning model .

[0012] In another implementation form of the first aspect , the modifying the first response comprises : classifying the first response using a second machine learning model and / or using at least one classification keyword in the first response , and modi fying the firstresponse based on the classification of the first response ; removing at least one section of the first response ; and / or replacing at least one section of the first response . The method can, for example , efficiently improve the response provided by the machine learning model .

[0013] In another implementation form of the first aspect , the providing the text input and the context of the text input to the machine learning model comprises : obtaining a modified text input by modifying the text input based at least on the context of the text input ; and providing the modified text input to the machine learning model . The method can, for example , provide the context of the text input to the machine learning model with improved compatibility .

[0014] In another implementation form of the first aspect , the modifying the text input comprises : adding at least one keyword to the text input based at least on the context of the text input ; removing at least one section of the text input based at least on the context of the text input ; and / or replacing at least one section of the text input based at least on the context of the text input . The method can, for example , efficiently add the context of the text input into the text input .

[0015] In another implementation form of the first aspect , the obtaining the text input from the user comprises : obtaining a speech input from the user ; and converting the speech input into the text input using a speech-to-text conversion . The method can, for example,improve responses provided by the machine learning model to speech input from the user .

[0016] In another implementation form of the first aspect , the determining the context of the text input comprises : determining the context of the text input by identifying at least one context keyword in the text input ; and / or determining the context of the text input using a third machine learning model . The method can, for example , efficiently determine the context of the text input .

[0017] In another implementation form of the first aspect , the machine learning model comprises a large language model . The method can, for example , utilise features of the large language model while obtaining more relevant responses by utilising the context of the text input .

[0018] According to a second aspect , a computing device compri ses at least one processor and at least one memory including computer program code , the at least one memory and the computer program code being configured to , with the at least one proces sor, cause the computing device to perform the method according to the first aspect .

[0019] According to a third aspect , a computer program product comprises program code configured to perform the method according to the first aspect when the computer program product is executed on a computer .

[0020] Many of the attendant features wil l be more readily appreciated as they become better understood byreference to the following detailed description considered in connection with the accompanying drawings .DESCRIPTION OF THE DRAWINGS

[0021] In the following, example embodiments are described in more detail with reference to the attached figures and drawings , in which :

[0022] Fig . 1 illustrates a flow chart representation of a method according to an embodiment ;

[0023] Fig . 2 illustrates a schematic representation of a computing device according to an embodiment ;

[0024] Fig . 3 illustrates a schematic representation of a data flow according to an embodiment ;

[0025] Fig . 4 illustrates a schematic representation of a data flow according to another embodiment ; and

[0026] Fig . 5 illustrates a schematic representation of a data flow according to another embodiment .

[0027] In the following, like reference numerals are used to des ignate li ke parts in the accompanying drawings .DETAILED DESCRIPTION

[0028] In the following description, reference is made to the accompanying drawings , which form part of the disclosure , and in which are shown, by way of illustration, specific aspects in which the present disclosure may be placed . It is understood that other aspects may be utilised, and structural or logical changes may bemade without departing from the scope of the present disclosure . The following detailed description, therefore , is not to be taken in a limiting sense , as the scope of the present disclosure is defined by the appended claims .

[0029] For instance , it is understood that a disclosure in connection with a described method may also hold true for a corresponding device or system configured to perform the method and vice versa . For example , if a specific method step is described, a corresponding device may include a unit to perform the described method step, even if such unit is not explicitly described or il lustrated in the f igures . On the other hand, for example , if a specific apparatus is described based on functional units , a corresponding method may include a step performing the described functionality, even if such step is not explicitly described or illustrated in the figures . Further, it is understood that the features of the various example aspects described herein may be combined with each other, unless specifically noted otherwise .

[0030] Fig . 1 illustrates a flow chart representation of a method according to an embodiment .

[0031] According to an embodiment , a computer-implemented method 100 for processing a text input from a user comprises obtaining 101 a text input from a user .

[0032] The text input may also be referred to as a request , a question, an input , or similar .

[0033] The text input may be obtained in various ways such as those di sclosed in the embodiments herein . For example , the user may be using a chat interface provided by a system, such as a server or a plurality of servers , implementing the method 100 or the user may send the text input via email , text message , or some other means .

[0034] The text input may comprise , for example , a question the user asking via a chat interface in, for example , a customer service situation .

[0035] The method 100 may further comprise determining102 a context of the text input .

[0036] The context of the text input may refer to any data and / or information that can further provide more information about the text input . For example, the context of the text input may comprise any data / inf ormation disclosed in the embodiments herein .

[0037] The method 100 may further comprise providing103 the text input and the context of the text input to a machine learning (ML) model .

[0038] The providing 103 the text input and the context of the text input to the ML model may comprise , for example , modifying the text input based on the context of the text input and providing the modified text input to the ML model and / or providing the text input and the context of the text input to the ML model separately . How the text input and the context of the text input are provided to the ML model may depend on, for example , the implementation of the ML model and / or what types of data input the ML model supports .

[0039] The method 100 may further comprise obtaining104 a first response from the machine learning model .

[0040] The first response may al so be referred to as first output , first ML output , or similar .

[0041] The ML model may provide the first response as a response to the text input and the context of the text input . For example , the text input and the context of the text input can be input into the ML model and the ML model may provide the first response as an output in response to that input .

[0042] The method 100 may further comprise providing105 a second response to the user based at least on the first response .

[0043] The providing 105 the second response to the user based at least on the first response may comprise , for example , providing the first response to the user as the second response and / or performing some processing operations , such as those disclosed herein, on the first response and providing the result of the processing to the user as the second response .

[0044] Herein, "obtaining" may comprise , for example , obtaining the data in question from memory, performing some processing and obtaining the data as a result of the processing, receiving the data from a func- tion / method / device / module / interf ace , reading a file containing the data, receiving the data via a data communication network, and / or similar .

[0045] Herein, some disclosure may be described in terms of functionality of a system . A system may referto any system configured to perform the method 100 . Any disclosure in relation to such a system can also be applied to the method 100 and vice versa .

[0046] The operations of the method 100 disclosed herein may be performed in various orders unless the opposite is stated explicitly or implicitly .

[0047] The utili zation of generic ML services can introduce various problems . For example , an ML service can be utili zed to automate customer service features , such as a customer service chat or customer service over the phone . For example , if a customer contacts the customer service of a specific mobile network operator and asks , "How do I order a phone contract ?" , a generic ML model may answer the question without taking into account that the customer has contacted the customer service of a specific mobile network operator, since the generic ML model does not have the full context of the question .

[0048] The method 100 can improve the responses provided by the ML model by specifying the context of the text input . This can be more eff icient than, for example , training and / or retraining an ML model using application-specific training data .

[0049] The method 100 can, for example utilise the ML model in a modular fashion . The implementation of the ML model can be independent of the method 100 and / or the device implementing the method 100 .

[0050] Fig . 2 illustrates a schematic representation of a computing device according to an embodiment .

[0051] According to an embodiment, a computing device 200 comprises at least one processor 201 and at least one memory 202 including computer program code, the at least one memory 202 and the computer program code configured to, with the at least one processor 201, cause the computing device 200 to perform the method 100.

[0052] The computing device 200 may comprise at least one processor 201. The at least one processor 201 may comprise, for example, one or more of various processing devices, such as a co-processor, a microprocessor, a digital signal processor (DSP) , a processing circuitry with or without an accompanying DSP, or various other processing devices including integrated circuits such as, for example, an application specific integrated circuit (ASIC) , a field programmable gate array (FPGA) , a microprocessor unit (MCU) , a hardware accelerator, a special-purpose computer chip, or the like.

[0053] The computing device 200 may further comprise a memory 202. The memory 202 may be configured to store, for example, computer programs and the like. The memory 202 may comprise one or more volatile memory devices, one or more non-volatile memory devices, and / or a combination of one or more volatile memory devices and nonvolatile memory devices. For example, the memory 202 may be embodied as magnetic storage devices (such as hard disk drives, magnetic tapes, etc.) , optical magnetic storage devices, and semiconductor memories (such as mask ROM, PROM (programmable ROM) , EPROM (erasable PROM) , flash ROM, RAM (random access memory) , etc.) .

[0054] The computing device 200 may further comprise other components not illustrated in the embodiment of Fig . 2 . The computing device 200 may comprise , for example , an input / output bus for connecting the computing device 200 to other devices .

[0055] When the computing device 200 is configured to implement some functionality, some component and / or components of the computing device 200 , such as the at least one processor 201 and / or the memory 202 , may be configured to implement this functionality . Furthermore , when the at least one processor 201 is configured to implement some functionality, this functionality may be implemented using program code comprised, for example , in the memory .

[0056] The computing device 200 may be implemented at least partially using, for example , a computer, some other computing device , or similar .

[0057] The method 100 and / or the computing device 200 may be utilised in, for example , automatic speech recognition (ASR) applications such as in a so-called voice- bot . The method 100 may be performed by, for example , a so-called chatbot that can be configured to interact with a user via text and / or a voicebot that can be configured to interact with a user via speech . The chat- bot / voicebot can also perform other processing / ac- tions / operations before and / or after performing the method 100 .

[0058] Fig . 3 illustrates a schematic representation of a data flow according to an embodiment .

[0059] In the embodiment of Fig . 3 , the user 301 can provide a text input 311 to the computing device 200 . The computing device 200 can provide the text input and the context of the text input 312 to a machine learning (ML) model 302 . The ML model can provide the first response 313 to the computing device 200 . The computing device 200 can provide the second response 314 to the user 301 .

[0060] In some embodiments , the ML model 302 may be provided by, for example , a separate service . For example , the ML model 302 may comprise a large language model (LLM) , such as generative pre-trained transformer 3 (GPT-3 ) or GPT-4 , provided via , for example ChatGPT . In other embodiments , the ML model 302 may be implemented by, for example , the computing device 200 or by some other computing device , such as a server .

[0061] The computing device 200 may communicate with the user 301 and / or with the ML model 302 via, for example , any communication channel , such as the Internet , Ethernet , Wi-Fi , a cellular network, and / or similar .

[0062] According to an embodiment , the determining the context of the text input comprises : obtaining metadata about the text input ; and determining the context of the text input based at least on the metadata about the text input .

[0063] The metadata can be obtained by, for example , identifying an identifying feature of the text input and obtaining the metadata based on the identifying feature .For example , the identifying feature may comprise the text input being provided by a specific user and the metadata can be obtained based on the specific user, such as identi fier of the user . Alternatively or addi tionally, the metadata may be obtained in various other ways such as those disclosed herein .

[0064] The metadata may be stored, for example , on the computing device 200 , such as in the memory 202 or in some other fashion, such as in a database the computing device 200 can access .

[0065] According to an embodiment , the metadata comprises static metadata .

[0066] The static metadata may also be referred to as fixed metadata or similar .

[0067] According to an embodiment , the static metadata comprises at least one of : information about a service the user is providing the text input for ; and / or information about a service the user is using to provide the text input .

[0068] The information about a service the user is providing the text input for may, for example , indicate that the text input is provided via a chat interface started on a webs ite of a specif ic product and / or service . The information about a service the user is providing the text input for may, for example , indicate that the text input corresponds to an email sent to a specific email address , such as an email address corresponding to a specific service , such as domestic flights of an airline . The information about a service the useris providing the text input for may, for example , indicate that the text input corresponds to a voice call whereby the user is calling a specific customer service number, such as loan customer service number of a bank .

[0069] The information about a service the user is using to provide the text input may, for example , indicate that the text input is provided via a chat interface , the text is received via email , the text input corresponds to a specific online form and / or a specific field on an online form, and / or the text input corresponds to a voice call .

[0070] For example , if the user is contacting the customer service of a mobi le network operator , the static metadata may indicate which mobile network operator the user is contacting . Since the method 100 may be implemented by the mobile network operator, this metadata may not change .

[0071] According to an embodiment , the metadata comprises dynamic metadata .

[0072] The dynamic metadata may also be referred to as changing metadata or similar .

[0073] According to an embodiment , the dynamic metadata comprises at least one of : time of day of obtaining the text input ; date of obtaining the text input : day of week of obtaining the text input ; time of year of obtaining the text input ; weather information ; status and / or error information ; and / or metadata relating to the user .

[0074] The status and / or error information may comprise , for example , information about disruptions in services . The method 100 can obtain this information from, for example , disturbance notices of an organi zation .

[0075] The metadata relating to the user can be obtained from, for example , a customer relationship management (CMR) system based on, for example , the telephone number of the user when the user contacts customer service via a voice cal l . The metadata relating to the user can indicate, for example, which services the user has ordered .

[0076] According to an embodiment , the machine learning model comprises a large language model .

[0077] LLMs can respond to various types of questions in the text input . However, since LLMs are typically trained with large quantities of generic data, providing further context for the text input can improve the response provided my LLMs .

[0078] According to an embodiment , the providing the text input and the context of the text input to the machine learning model comprises : obtaining a modified text input by modifying the text input based at least on the context of the text input ; and providing the modified text input to the machine learning model .

[0079] According to an embodiment , the modifying the text input comprises : adding at least one keyword to the text input based at least on the context of the text input ; removing at least one section of the text inputbased at least on the context of the text input ; and / or replacing at least one section of the text input based at least on the context of the text input .

[0080] The adding at least one keyword to the text input based at least on the context of the text input may comprise, for example , adding such a keyword to the text input so that the ML model provides a more appropriate response for the context .

[0081] For example , if text input comprises the question "How do I order a mobile phone plan?" presented by the user in a customer service chat of a mobile network operator, the adding at least one keyword to the text input may comprise adding the name of the mobile network operator as the keyword to the text input . Thus , the ML model may be able to provide more accurate response for this context .

[0082] For example , the text data can comprise the question "How do I complain about a fault ?" provided by the user in a customer service situation to a mobile network operator . This question can be replaced with "fault reports" .

[0083] In some embodiments , the context of the text input may be provided separately from the text input to the ML model . For example , if the ML model supports providing additional information in addition to the text input , such as a topic and / or classification of the text input , the context of the text input can be provided as such additional information .

[0084] According to an embodiment , the determining the context of the text input compri ses : determining the context of the text input by identifying at least one context keyword in the text input ; and / or determining the context of the text input using a third machine learning model .

[0085] For example , the method 100 may determine based on the at least one context keyword in the text input and / or using the third machine learning model that the text input relates to a fault in a service . The service fault can then be provided as the context of the text input to the ML model .

[0086] The third ML model may comprise , for example , any ML model that has been trained to determine the context of the text input .

[0087] According to an embodiment , the providing the second response to the user based at least on the first response comprises : modifying the first response ; and providing at least the modified first response to the user in the second response .

[0088] According to an embodiment , the modifying the first response comprises : classifying the first response using a second machine learning model and / or using at least one classification keyword in the first response , and modifying the first response based on the classification of the first response ; removing at least one section of the first response ; and / or replacing at least one section of the first response .

[0089] The replacing at least one section of the first response may comprise , for example , replacing one or more words / sentences in the first response or replacing the whole first response .

[0090] The removing at least one section of the first response may comprise , for example, removing irrelevant parts of the first response .

[0091] The removing at least one section of the first response may comprise, for example removing one or more words / sentences in the first response . For example , the classification may indicate that a sentence in the first response is irrelevant . Such a sentences may then be removed .

[0092] For example , if text input comprised the question "How do I order a mobile phone plan?" presented by the user in a customer service chat of a mobile network operator and the first response comprises "You can order a mobile plan by contacting the mobile network operator . " , modifying the f irst response based on the clas sification of the first response may comprise replacing the section "the mobile network operator" with the name of the mobile network operator the customer service chat of which the user is contacting .

[0093] Fig . 4 illustrates a schematic representation of a data flow according to another embodiment .

[0094] In the embodiment of Fig . 4 , the computing device 200 is used to implement a voicebot 401 and a context module 402 . The voicebot 401 can communicatewith the user 301 via audio 411 . For example , the voice- bot 401 can communicate with the user 301 via a voice call . The voice call can comprise , for example , a tel ephone call , a voice over internet protocol (VoI P) call , or any other type of voice call .

[0095] For example , the voicebot 401 may obtain a speech input from the user 301 and convert the speech input into the text input using a speech-to-text conversion . The text input can then be used to perform any processing operations , such as those disclosed herein . The voicebot 401 may also convert the second response to speech using a text-to-speech conversion . The speech can then be provided to the user 301 via audio 411 .

[0096] According to an embodiment , the obtaining 101 the text input from the user comprises : obtaining a speech input from the user ; and converting the speech input into the text input using a speech-to-text conversion .

[0097] The context module 402 may be configured to , for example , determine the context of the text input , modify the text input , and / or modify the first response . The context module 402 can communicate 412 with the voicebot 401 to , for example , obtain the text input from the voicebot 401 and to provide the second response to the voicebot 401 . The context module 402 may also communicate 413 with the ML model 302 to provide the text input and the context of the text input to the ML model 302 and to obtain the f irst response from the ML model302 .

[0098] Fig . 5 illustrates a schematic representation of a data flow according to another embodiment .

[0099] The embodiment of Fig . 5 illustrates an example of how the method 100 can be implemented .

[0100] The text input 311 , static metadata 502 , and dynamic metadata 503 can be input into a classification module 504 . The classification module 504 can output information about classification of the text input 311 . The text input 311 may comprise , for example , a question and / or a request presented by the user .

[0101] The text input 311 , static metadata 502 , dynamic metadata 503 , and the information about the classification of the text input 311 can be input into a context determination module 505 . The context determination module 505 can determine the context of the text input 311 based on the aforementioned data / inf ormation .

[0102] A text input modification module 506 can modify the text input 311 based on the context of the text input 311 .

[0103] The text input and the context of the text input can be transmitted 507 to the ML model . The first response can be modified 508 to obtain the second response .

[0104] Any range or device value given herein may be extended or altered without losing the effect sought . Also any embodiment may be combined with another embodiment unless explicitly disallowed .

[0105] Although the subj ect matter has been described in language specific to structural features and / or acts ,it is to be understood that the subj ect matter defined in the appended claims is not necessarily limited to the specific features or acts described above . Rather, the specific features and acts described above are disclosed as examples of implementing the claims and other equivalent features and acts are intended to be within the scope of the claims .

[0106] It will be understood that the benefits and advantages described above may relate to one embodiment or may relate to several embodiments . The embodiments are not limited to those that solve any or all of the stated problems or those that have any or all of the stated benefits and advantages . It wil l further be understood that reference to ' an ' item may refer to one or more of those items .

[0107] The steps of the methods described herein may be carried out in any suitable order, or simultaneously where appropriate . Additionally, individual blocks may be deleted from any of the methods without departing from the spirit and scope of the subj ect matter described herein . Aspects of any of the embodiments described above may be combined with aspects of any of the other embodiments described to form further embodiments without losing the effect sought .

[0108] The term ' comprising ' is used herein to mean including the method, blocks or elements identified, but that such blocks or elements do not comprise an exclusive list and a method or apparatus may contain additional blocks or elements .

[0109] It will be understood that the above description is given by way of example only and that various modif ications may be made by those ski lled in the art . The above specification, examples and data provide a complete description of the structure and use of exemplary embodiments . Although various embodiments have been described above with a certain degree of particularity, or with reference to one or more individual embodiments , those skilled in the art could make numer- ous alterations to the disclosed embodiments without departing from the spirit or scope of this specification .

Claims

CLAIMS :

1. A computer-implemented method (100) for processing a text input from a user, the method comprising : obtaining (101) a text input from a user; determining (102) a context of the text input by obtaining metadata about the text input and determining the context of the text input based at least on the metadata about the text input; providing (103) the text input and the context of the text input to a machine learning model, wherein the machine learning model comprises a large language model ; obtaining (104) a first response from the machine learning model; and providing (105) a second response to the user based at least on the first response.

2. The computer-implemented method (100) according to claim 1, wherein the metadata comprises static metadata.

3. The computer-implemented method (100) according to claim 2, wherein the static metadata comprises at least one of: information about a service the user is providing the text input for; and / orinformation about a service the user is using to provide the text input.

4. The computer-implemented method (100) according to any of claims 1 - 3, wherein the metadata comprises dynamic metadata.

5. The computer-implemented method (100) according to claim 4, wherein the dynamic metadata comprises at least one of: time of day of obtaining the text input; date of obtaining the text input; day of week of obtaining the text input; time of year of obtaining the text input; weather information; status and / or error information; and / or metadata relating to the user.

6. The computer-implemented method (100) according to any preceding claim, wherein the provision (105) of the second response to the user based at least on the first response comprises: modifying the first response; and providing at least the modified first response to the user in the second response.

7. The computer-implemented method (100) according to claim 6, wherein the modifying the first response comprises:classifying the first response using a second machine learning model and / or using at least one clas sification keyword in the first response , and modifying the first response based on the classification of the first response ; removing at least one section of the first response ; and / or replacing at least one section of the first response .8 . The computer-implemented method ( 100 ) according to any preceding claim, wherein the providing ( 103 ) the text input and the context of the text input to the machine learning model comprises : obtaining a modified text input by modifying the text input based at least on the context of the text input ; and providing the modified text input to the machine learning model .9 . The computer-implemented method ( 100 ) according to claim 8 , wherein the modifying the text input comprises : adding at least one keyword to the text input based at least on the context of the text input ; removing at least one section of the text input based at least on the context of the text input ; and / or replacing at least one section of the text input based at least on the context of the text input .

10. The computer-implemented method (100) according to any preceding claim, wherein the obtaining(101) the text input from the user comprises: obtaining a speech input from the user; and converting the speech input into the text input using a speech-to-text conversion.

11. The computer-implemented method (100) according to any preceding claim, wherein the determining(102) the context of the text input comprises: determining the context of the text input by identifying at least one context keyword in the text input; and / or determining the context of the text input using a third machine learning model.

12. A computing device (200) , comprising at least one processor (201) and at least one memory (202) including computer program code, the at least one memory (202) and the computer program code configured to, with the at least one processor (201) , cause the computing device (200) to perform the method (100) according to any preceding claim.

13. A computer program product comprising program code configured to perform the method (100) according to any of claims 1 - 11 when the computer program product is executed on a computer.

Images