Techniques for efficient retrieval of personality data

Abstract

Techniques for efficient retrieval of personality data are disclosed. A technique for enabling a client device (406) to efficiently retrieve a digital representation of personality data of a user (402) from a server (404), wherein the digital representation of personality data is processed at the client device (406) to provide a user-adapted service to the user (402). A method implementation of the technique is performed by the server (404) and includes: storing a neural network trained to compute personality data of a user based on input obtained from the user; receiving, from the client device (406), a request for a digital representation of personality data of the user (402); and sending, to the client device (406), the requested digital representation of personality data of the user (402), wherein the personality data of the user is computed using the neural network based on input obtained from the user.

Description

[0001] This application is a divisional application of Chinese invention patent application filed on March 18, 2020, with application number 2020800221084 and title "Technology for Efficient Retrieval of Personality Data". Technical Field

[0002] This disclosure generally relates to the field of data retrieval. In particular, a technique is presented for enabling efficient retrieval of a user's personality data from a server by a client device. This technique can be implemented in the form of methods, computer programs, apparatus, and systems. Background Technology

[0003] For decades, personality tests have been used to assess personality traits and are typically performed based on personality survey data obtained from test subjects, which is then evaluated by professionals such as psychologists to draw conclusions about the individual's personality. The so-called "OCEAN" model is a widely accepted classification of personality traits, also known as the "Big Five" personality traits, and includes openness, conscientiousness, extraversion, agreeableness, and neuroticism as personality dimensions. Well-known personality tests using the OCEAN model include those based on the so-called International Personality Item Pool (IPIP), the HEXACO-60 scale, and the Big-Five-Inventory-10 (BFI-10), which consist of a set of questions designed to test a person on each of the five personality dimensions. However, because conventional personality tests typically require review by human specialists such as psychologists to obtain a qualified assessment of a person's personality traits, it is difficult to integrate the implementation of personality tests and their results into the process of execution on a technological system, even though such integration could be beneficial as it would allow for a better fit to the user's personality and thus improve the user experience, such as by providing users with user-fitted services. Summary of the Invention

[0004] Therefore, there is a need for a technical implementation method that makes it practically feasible to integrate personality tests and their results into the process of execution on a technical system.

[0005] According to a first aspect, a method is provided for enabling a client device to efficiently retrieve a digital representation of a user's personality data from a server, wherein the digital representation of the personality data is processed at the client device to provide a user-tailored service to the user. The method is executed by a server and includes: storing a neural network trained to compute the user's personality data based on input obtained from the user; receiving a request for a digital representation of the user's personality data from the client device; and sending the requested digital representation of the user's personality data to the client device, wherein the user's personality data is computed using the neural network based on input obtained from the user.

[0006] By storing a trained neural network on a server and applying it to calculate a user's personality data, the retrieval of the digital representation of the user's personality data can be automated (as routine manual review is no longer required). Therefore, integrating the retrieval and use of user personality data into processes executed on a technical system (e.g., automated processes) becomes feasible. In particular, a neural network can be viewed as an efficient functional data structure capable of computing the requested personality data in a single computational run; that is, by inputting input obtained from the user at the neural network's input nodes and reading the resulting output values ​​representing the personality data from the neural network's output nodes. Thus, a neural network can efficiently provide personality data in digital representation to client devices, which can be used to provide services tailored to the user's specific personality, thereby improving the user experience on the client device side. The integration of personality data retrieval and use becomes particularly practical due to the efficient provision of data, as the digital representation of personality data can be provided to client devices without significant latency and can be processed immediately at the client device. Therefore, a technical implementation can be implemented that substantially makes it practically feasible to integrate the retrieval and use of personality data into processes executed on a technical system.

[0007] User personality data can indicate a user's psychological characteristics and / or preferences, and thus, personality data can typically include psychological and medical data (e.g., data indicating tendencies such as curiosity, anxiety, and depression), including classic personality data that can be based, for example, on personality dimensions such as openness, conscientiousness, extraversion, agreeableness, and neuroticism (as mentioned above, known as the Big Five personality traits). The numerical representation of a user's personality data can include numerical representations of the mentioned traits, such as numerical representations of at least one of the personality dimensions of openness, conscientiousness, extraversion, agreeableness, and neuroticism calculated for the user by a neural network.

[0008] Client devices can be configured to process digital representations of personality data to enable the provision of user-tailored services. In one variant, the client device itself can be configured based on a digital representation of personality data. For example, an exemplary device that can be configured via a digital representation of personality data could be a vehicle. In this case, the vehicle could be the client device. The vehicle can process the received digital representation of a user's (e.g., the vehicle's driver's) personality data and configure itself (e.g., including its sub-components) to adapt the vehicle's driving configuration to the driver's personality, thereby providing driving services specifically tailored to the user's personality. For example, if the personality data indicates that the driver tends to be risk-averse or anxious, the vehicle's driving configuration can be configured to be more safety-oriented, while for a driver tending to have a more risk-seeking personality, the vehicle's driving configuration can be configured to be more sporty. For this purpose, the vehicle's throttle and braking response behavior can be adapted accordingly in other settings. Sub-components of the vehicle providing vehicle-related services can also be configured based on personality data, such as the vehicle's sound system, including its voice and volume settings, to better match the user's personality, for example.

[0009] In another variation, the client device can configure at least one other device based on a digital representation of personality data, for example, when it is at least one other device providing services to the user. In this variation, the client device can be, for example, a mobile terminal (e.g., a smartphone) that can interface with a vehicle (e.g., using Bluetooth) (i.e., in this case, the vehicle corresponds to the at least one other device), and the mobile terminal can configure the vehicle via the interface when receiving the digital representation of personality data from the server. Therefore, it can be said that the digital representation of the user's personality data can be processed at the client device to configure at least one device providing services to the user. Configuring at least one device may include configuring at least one setting of the at least one device and / or configuring at least one setting of the services provided by the at least one device. It should be understood that a vehicle is merely an example of a device that can be configured based on personality data, and the client device and / or at least one other device may also correspond to other types of devices.

[0010] In one implementation, the method performed by the server may further include: receiving feedback characterizing the user; updating the neural network based on the feedback; and sending a digital representation of the user's updated personality data to a client device, wherein the updated neural network can be used to compute the user's updated personality data. The digital representation of the user's updated personality data may be processed at the client device to improve the configuration of at least one device providing services to the user (e.g., one of the configurations of the vehicles mentioned above). Feedback may be collected at the client device and / or at least one device providing services to the user and may indicate the user's personality. For example, the feedback may include behavioral data reflecting the behavior of a user monitored at said at least one device when using services provided by said at least one device, wherein, in a variant, the behavioral data may be monitored using measurements performed by the at least one device providing services to the user (e.g., sensor-based measurements). In the vehicle example, for example, the monitored user's behavior may be the user's driving behavior, and driving behavior may be measured by sensors at the vehicle. For example, to measure driving behavior, sensors may sense the user's braking response and intensity, and since such measurements may indicate the user's personality (e.g., aggressiveness in driving), this information may be sent as feedback to the server to update the neural network and thereby improve the neural network's ability to compute the user's personality data.

[0011] Updating a neural network can include training it based on feedback received from a client device. If the feedback represents a new input value that has not yet been input to the neural network, a new input node can be added, and the new input value can be assigned to the new input node during training. This makes the neural network's ability to function as an effective data structure in the technical implementation presented herein particularly evident: the neural network represents an effective, updatable data structure that can be updated based on arbitrary feedback received from the client device regarding the user's personality to improve its ability to compute personality data. The information conveyed by the feedback can be directly integrated into the neural network and, once trained, can be immediately reflected in subsequent requests for a digital representation of the requested personality data sent to the server. Conventional personality assessment techniques are rather fixed and may not support this updability at all.

[0012] The digital representation of a user's personality sent from the server to a client device can correspond to a digital representation of the user's personality previously calculated by the server in a previous request to calculate the user's personality (e.g., when performing a personality test by having the user answer a set of questions). Therefore, the user's personality data can be calculated before receiving a request from the client device, where the request may include an access code previously provided to the user by the server when calculating the user's personality data, allowing the user to access the digital representation of the user's personality data from different client devices. This implementation saves computational resources at the server because the digital representation of the user's personality does not need to be recalculated each time a digital representation of a specific user's personality data is requested from a client device; instead, it can be returned based on pre-calculated personality data. Conversely, the user can use the access code to access the digital representation of personality data from multiple different client devices, such as different vehicles the user can drive, such as cars and motorcycles, or other types of devices.

[0013] The input obtained from the user can correspond to numerical scores (e.g., obtained in a question-and-answer scheme as a personality test) reflecting answers to questions about at least one of the user's personality, purpose, and motivation. Each numerical score can be used as input to a separate input node of the neural network when calculating the user's personality data using a neural network. For example, the numerical scores can correspond to a five-point Likert scale with values ​​from 1 to 5. The neural network can correspond to a deep neural network with at least two hidden layers between an input layer including the input nodes and an output layer including the output nodes of the neural network. For example, personality-related questions can correspond to questions from the standard IPIP, HEXACO-60, and / or BFI-10 pools; however, it should be understood that other questions about the user's personality, including questions about the user's psychological characteristics and / or preferences, can also be used. Questions specifically related to the user's purpose and motivation can define additional dimensions (e.g., in addition to the Big Five personality traits), which can improve the accuracy of the calculated personality data compared to standard IPIP, HEXACO-60, and BFI-10 techniques. The network can be trained based on data collected from a basic survey conducted by multiple testers (e.g., 1,000 or more), where the aforementioned questions can be used to conduct the basic survey.

[0014] To reduce computational complexity when calculating a user's personality data, neural networks can be designed with specific network structures. Given the scenario described above, the structure of a neural network can typically be designed such that the number of input nodes is reduced compared to the number of input nodes available when using all of the aforementioned problems. Therefore, the problem can correspond to a question selected from a set of questions representing the best achievable outcome for calculating the user's personality data (i.e., if all questions in the set are answered by the user), where the selected question corresponds to the question determined to be most influential relative to the best achievable outcome. As mentioned above, since each answer to a question can be input to a separate input node of the neural network, selecting a subset of the question set reduces the number of input nodes when calculating personality data, thereby reducing computational complexity. Because of the fact that the question most influential relative to the achievable outcome is selected, the accuracy of the results output by the neural network can be largely maintained.

[0015] In fact, tests have shown that the number of questions can be significantly reduced without significantly sacrificing the accuracy of the results. Using a set of questions comprising standard IPIP, HEXACO-60, and BFI-10 questions (a total of 370 questions) (optionally supplemented by additional questions about the user's purpose and motivation, bringing the total to over 370 questions) as the optimal set for calculating personality data, tests have shown that approximately 90% accuracy of the optimal set can be achieved when only 30 of the most influential questions are used. Therefore, the number of selected questions can be less than 10% (preferably less than 5%) of the number of questions included in the optimal set. This is because, in this case, the number of input nodes to the neural network can be significantly reduced, resulting in substantial savings in computational resources and more efficient calculation of personality data.

[0016] To determine the problems in the set that have the greatest impact relative to the best achievable outcome, in one variation, problems can be selected from the set based on associating the achievable outcome of each individual problem in the set with the best achievable outcome and choosing the problem from the set that has the highest correlation with the best achievable outcome. Therefore, a fixed subset of the set representing the best achievable outcome can be determined, and then, as described above, this subset can be used to train a neural network with a reduced number of input nodes.

[0017] As described, the best achievable result can correspond to the result achieved when the user answers all questions in the question set (such as a question set including standard IPIP, HEXACO-60, and BFI-10 questions, optionally supplemented by additional questions about the user's goals and motivations). While in one variation, the standard IPIP score (obtained by answering all questions in the standard IPIP test), the standard HEXACO-60 score (obtained by answering all questions in the standard HEXACO-60 test), and the standard BFI-10 score (obtained by answering all questions in the standard BFI-10 test) can each serve as a reference to the best achievable result individually, in another variation, an improvement can be achieved by calculating a combined score of these individual scores as a reference to the best achievable result, where the combined score can be calculated, for example, as the average of the individual scores (e.g., a weighted average). The combined score can also be represented as a "super-score" representing the "truth value" that can be derived from the individual scores, substantially enhancing the meaning of the determined score and representing an improved reference to the best achievable result.

[0018] In another variation, questions can be iteratively selected from a set of questions, wherein in each iteration, the next question can be selected based on the user's answers to previous questions, and wherein, in each iteration, the next question can be selected as the question in the set that is determined to have the greatest impact on the achievable outcome for calculating the user's personality data. This can be viewed as adaptive question selection, where questions are determined stepwise and in a user-specific manner, taking into account the user's answers to previous questions. In a particular variation, the neural network can include multiple output nodes of a probability curve representing the outcome of the user's personality data, wherein determining the most influential question in the set as the next question for the corresponding iteration can include, for each input node of the neural network, determining to what extent a change in the numerical rating input to the corresponding input node alters the probability curve. The question associated with the input node in the probability curve whose degree of change is determined to be the highest can be selected as the most influential question for the corresponding iteration.

[0019] To further reduce computational complexity, the above iterations and adaptive selection can be performed under at least one constraint, such as the maximum number of questions to select, the minimum result accuracy to achieve (result accuracy can increase with each question answered in each iteration, and computation can stop when the desired minimum result accuracy is reached), and the maximum available time (the test can stop when the maximum available time has elapsed, or each question can be associated with an estimated time to be answered by the user, and the number of questions to select can be determined based on the estimated time). These constraints can be configured separately for each computation of the personality data.

[0020] According to a second aspect, a method is provided for enabling a client device to efficiently retrieve a digital representation of a user's personality data from a server. The method is performed by the client device and includes: sending a request to the server for a digital representation of the user's personality data; receiving the requested digital representation of the user's personality data from the server, the user's personality data being computed based on input obtained from the user using a neural network trained to compute the user's personality data based on the input obtained from the user; and processing the digital representation of the personality data to provide the user with a user-adapted service.

[0021] The method according to the second aspect defines a method from the perspective of the client device, which can be complementary to the method executed by the server according to the first aspect. The server and client device of the second aspect can correspond to the server and client device described above with respect to the first aspect. Therefore, those aspects described with respect to the method of the first aspect (which are applicable to the method of the second aspect) can also be included in the method of the second aspect, and vice versa. Therefore, unnecessary repetition is omitted below.

[0022] As in the method of the first aspect, a digital representation of the user's personality data may be processed at a client device to configure at least one device providing services to the user, wherein the at least one device may include the client device. The method performed by the client device may further include: sending feedback characterizing the user to a server; and receiving an updated digital representation of the user's personality data from the server, wherein the updated personality data may be computed using a neural network updated based on the feedback. The updated digital representation of the user's personality data may be processed at the client device to improve the configuration of the at least one device providing services to the user. The feedback may include behavioral data reflecting the user's behavior monitored at the at least one device when using services provided by the at least one device, wherein the behavioral data may be monitored using measurements performed by the at least one device providing services to the user. The at least one device may include a vehicle, wherein the behavioral data may include data reflecting the user's driving behavior. The user's personality data may be computed before sending a request to the server, wherein the request may include an access code previously provided to the user by the server when computing the user's personality data, the access code allowing the user to access the digital representation of the user's personality data from different client devices. Input obtained from the user may correspond to a digital score reflecting answers to questions about at least one of the user's personality, purpose, and motivation.

[0023] According to a third aspect, a computer program product is provided. The computer program product includes program code portions for performing at least one of the methods of the first and second aspects when the computer program product is executed on one or more computing devices (e.g., a processor or a distributed processor group). The computer program product can be stored on a computer-readable recording medium, such as semiconductor memory, DVD, CD-ROM, etc.

[0024] According to a fourth aspect, a server is provided for enabling a client device to efficiently retrieve a digital representation of a user's personality data from a server, wherein the digital representation of the personality data is processed at the client device to provide the user with a user-adapted service. The server includes at least one processor and at least one memory, wherein the at least one memory contains instructions executable by the at least one processor to enable the server to perform any method steps presented herein with respect to the first aspect.

[0025] According to a fifth aspect, a client device is provided for efficiently retrieving a digital representation of a user's personality data from a server. The client device includes at least one processor and at least one memory, wherein the at least one memory contains instructions executable by the at least one processor to enable the client device to perform any method steps presented herein with respect to a second aspect.

[0026] According to a sixth aspect, a system is provided, the system comprising a server according to a fourth aspect and at least one client device according to a fifth aspect. Attached Figure Description

[0027] Further details and advantages of the techniques presented herein will be described with reference to exemplary embodiments illustrated in the accompanying drawings, in which:

[0028] Figure 1a and Figure 1b Exemplary components of a server and client device according to this disclosure are shown;

[0029] Figure 2 The method that can be executed by a server according to this disclosure is shown;

[0030] Figure 3 A method according to this disclosure that can be executed by a client device is shown;

[0031] Figure 4 Exemplary interactions between a user, server, and client device (using a car as an example) according to this disclosure are illustrated;

[0032] Figure 5 Different connection options between a user's mobile terminal, a car, and a server are shown according to this disclosure;

[0033] Figure 6a and Figure 6b An exemplary structure of a neural network according to this disclosure is shown;

[0034] Figure 7 An exemplary embodiment according to this disclosure is shown, which involves adapting vehicle settings to take into account the driver's level of attention; and

[0035] Figure 8 An exemplary implementation according to this disclosure is shown, which involves taking into account a user's body scan data to provide the user with a user-adaptive service. Detailed Implementation

[0036] In the following description, specific details are set forth for purposes of explanation and not limitation in order to provide a thorough understanding of this disclosure. It will be apparent to those skilled in the art that this disclosure can be practiced in other ways departing from these specific details.

[0037] Those skilled in the art will further understand that the steps, services, and functions explained below can be implemented using separate hardware circuit systems, software running in conjunction with a programmed microprocessor or general-purpose computer, one or more application-specific integrated circuits (ASICs), and / or one or more digital signal processors (DSPs). It will also be understood that, when this disclosure is described in terms of method, it can also be implemented in one or more processors and one or more memories coupled to the one or more processors, wherein the one or more memories are encoded with one or more programs that, when executed by the one or more processors, perform the steps, services, and functions disclosed herein.

[0038] Figure 1a An exemplary configuration of a server 100 is schematically illustrated for enabling a client device to efficiently retrieve a digital representation of a user's personality data from the server 100, wherein the digital representation of the personality data is processed at the client device to provide the user with a user-tailored service. The server 100 includes at least one processor 102 and at least one memory 104, wherein the at least one memory 104 contains instructions that can be executed by the at least one processor 102, such that the requesting server 100 is operable to perform the method steps described herein with reference to "server".

[0039] It should be understood that server 100 can be implemented on physical computing units or virtual computing units (such as virtual machines). It should also be understood that server 100 does not necessarily have to be implemented on a separate computing unit, but can be implemented as a component (implemented as software and / or hardware) residing on multiple distributed computing units, such as in a cloud computing environment.

[0040] Figure 1b An exemplary composition of a client device 110 is schematically illustrated, enabling the client device 110 to efficiently retrieve a digital representation of a user's personality data from a server. The client device 110 includes at least one processor 112 and at least one memory 114, wherein the at least one memory 114 contains instructions that can be executed by the at least one processor 112, such that the requesting client device 110 is operable to perform the method steps described herein with reference to "client device".

[0041] Figure 2 A method according to this disclosure, which can be performed by server 100, is illustrated. This method aims to enable a client device (e.g., client device 110) to efficiently retrieve a digital representation of a user's personality data from server 100. In this method, server 100 can perform the steps described herein with reference to "server," and according to the above description, in step S202, server 100 can store a neural network trained to compute the user's personality data based on input obtained from the user; in step S204, server 100 can receive a request for a digital representation of the user's personality data from the client device; and in step S206, server 100 can send the requested digital representation of the user's personality data to the client device, wherein the user's personality data is computed using a neural network based on input obtained from the user.

[0042] Figure 3 A method according to this disclosure, which can be performed by a client device 110, is illustrated. This method aims to enable the client device 110 to efficiently retrieve a digital representation of a user's personality data from a server (e.g., server 100). In this method, the client device 110 can perform the steps described herein with reference to "client device," and according to the above description, in step S302, the client device 110 can send a request to the server for a digital representation of the user's personality data; in step S304, the client device 110 can receive from the server the requested digital representation of the user's personality data, the user's personality data being calculated based on input obtained from the user, using a neural network trained to calculate the user's personality data based on input obtained from the user; and in step S306, the client device 110 can process the digital representation of the personality data to provide the user with a user-adapted service.

[0043] Figure 4The illustration shows an exemplary interaction between a user 402, a server 404 storing a neural network trained to calculate the user's personality data based on input obtained from the user, and a client device for retrieving a digital representation of the user 402's personality data to provide the user 402 with a user-tailored service. In the example shown, the client device is a car 406 that can be driven by the user 402. As shown, the user 402 can perform an automated personality test by answering questions, for example, using a web interface or application on his laptop or smartphone, thereby providing input to the neural network stored at the server 404. Based on this input, the neural network can calculate the user 402's personality data. In the example shown, instead of sending a digital representation of the personality data, the server 404 provides the user 402 with an access code, which the user 402 can use to access the personality data using different client devices, including the car 406. User 402 can register or log in at vehicle 406 (more specifically, at its onboard computer) using an access code, and then vehicle 406 can use the access code to request a digital representation of the user's personality data from server 404 (in the figure, the user's personality data is represented as the user's "MindDNA").

[0044] Once a request is received from vehicle 406, server 404 can return the user's personality data to vehicle 406. The vehicle can then configure its driving configuration (and optionally, sub-components of vehicle 406) based on the user's personality data, for example, adapting the throttle and braking response behavior of vehicle 406 to provide a driving experience particularly suited to the user's personality (e.g., risk-averse, risk-seeking, etc.). When user 402 then drives vehicle 406, vehicle 406 can monitor the user's driving behavior, for example, using sensors that measure the user's braking response and intensity. Vehicle 406 can provide this information as feedback to server 404, where the feedback can be processed to update (through training) a neural network to improve its ability to calculate user 402's personality data. In response, server 404 can send the corresponding updated personality data of user 402 to vehicle 406, which can then use a digital representation of the updated personality data to improve its vehicle configuration to better align with the actual personality of user 402. In summary, a system is provided that allows the retrieval and use of user personality data to be integrated into automated processes to adapt the configuration of devices or services provided on them to user preferences derived from the user's personality data, thereby improving the user experience.

[0045] Figure 5Different connection options are illustrated between a user 402's mobile terminal 502 (e.g., a smartphone), a vehicle 406, and a server 404 according to this disclosure. In one variation, the vehicle 406 can communicate directly with the server 404 via the Internet, and when the user 402 authenticates with the vehicle 406 (e.g., using a key, smart card, NFC / RFID, an NFC-enabled smartphone, fingerprint, etc.), the vehicle 406 can request the user's personality data (in... Figure 5 This is again referred to as the user's "MindDNA" to improve the driving experience of user 402. In another variation, when user 402 carries mobile terminal 502, mobile terminal 502 can (e.g., using a dedicated application installed on it) communicate with server 404 via the Internet and request user 402's personality data. In this variation, vehicle 406 can communicate locally with mobile terminal 502 (e.g., using Bluetooth, Wi-Fi, or a USB cable) and retrieve the user's personality data from mobile terminal 502. The direct connection between vehicle 406 and mobile terminal 502 can be additionally used to supplement the feedback collected by vehicle 406 itself (e.g., related to the user's driving behavior) with sensors installed on mobile terminal 502 (e.g., a gyroscope for motion and acceleration detection, GPS for motion and acceleration detection and driving route detection, or medical sensors for measuring pulse, blood pressure, etc.), thereby providing server 404 with additional feedback sensed by mobile terminal 502 for updating the neural network based on this feedback, as described above.

[0046] Figure 6a An exemplary structure of a neural network 602 according to this disclosure is shown. The neural network 602 includes an input layer, an output layer, and two hidden layers. It should be understood that... Figure 6aThe neural network 602 shown only illustrates the general structure of a deep neural network, and the actual number of nodes in the neural network 602 stored in server 404 (at least in the input and hidden layers) can be significantly higher than the number shown in the figure. As mentioned above, the neural network 602 has been tested with 30 of the most influential questions from a total of 370 or more questions (taken from standard IPIP, HEXACO-60, and BFI-10 questions, and optionally supplemented by further questions about the user's purpose and motivation), resulting in 30 input nodes in the input layer of the neural network 602. In this case, for example, each of the hidden layers can be configured with 50 nodes. Furthermore, as shown, the neural network 602 can include a single output node in the output layer. In this case, the result value at the output node of the output layer can represent a value of a personality dimension (one of the five major personality traits) on which the neural network 602 has been trained. It should be understood that this structure of the neural network 602 is merely exemplary, and other structures are generally conceivable.

[0047] A more advanced structure of the neural network 602 includes input nodes based on the number of the entire set of available questions, which can be taken from standard IPIP, HEXACO-60, and BFI-10 questions, including additional questions about the user's purpose and motivation, and further additional questions about other psychological characteristics and / or preferences of the user that are not covered by the above questions, potentially totaling hundreds of questions, such as more than 600 questions. Such a neural network 602 can therefore have more than 600 input nodes, each corresponding to one of the questions in the entire set of available questions, and the number of nodes in the hidden layers can be selected based on the performance of the neural network 602. For example, the neural network 602 may include two hidden layers, each with 100 nodes. Further, in the input layers, the aforementioned more than 600 input nodes can be duplicated, where each duplicated input node can serve as a missing question indicator. Missing question indicators can be binary, that is, they can have only two values ​​(e.g., 0 and 1) indicating whether the question of the corresponding (original) input node has been answered. Due to the duplicated input nodes, the input layers can include a total of more than 1200 input nodes.

[0048] The output layer of a more advanced neural network 602 can have multiple output nodes that together represent the probability curve of a personality dimension. For example, if the scale used for the output in this personality dimension ranges from 0 to 10, and the number of output nodes is 50, then each output node can represent a portion of the scale, corresponding to different parts of the scale: 0 to 0.2, 0.2 to 0.4, 0.4 to 0.6, ..., 9, 8, 10. Instead of a single output value, such an output layer can deliver the entire probability curve of the output values ​​for this personality dimension. Figure 6b An exemplary output layer and a corresponding probability curve 604 are shown. This curve allows for determining where the output value is most likely to be (i.e., indicated by the peak of the curve) and the accuracy of the computation results of the neural network 602 (i.e., indicated by the width of the curve). Using the advanced neural network 602, assuming that the neural network 602 is trained separately for each dimension, the user's personality data can be computed in the form of several probability curves (e.g., five probability curves corresponding to the five major personality traits) for any number of answered questions. In the initial state where no questions have been answered, the values ​​of all missing question indicators can be "missing" (e.g., 0). With each subsequent answer to a question, an update to the output value can be computed such that the width of the probability curve on the output layer decreases as the number of questions answered increases, resulting in a steady increase in the accuracy of the computation results of the neural network 602.

[0049] This structure of neural network 602 can be particularly advantageous because it allows for iterative selection of the next question to be answered by the user from the entire question set. In each iteration, the next question can be selected based on the user's answers to previous questions, specifically the question identified as having the greatest impact on the achievable outcome for calculating the user's personality data. To this end, several (e.g., five) probability curves can be recalculated for each answered question, and the one with the largest width (i.e., representing the current lowest accuracy) can be determined from among the recalculated probability curves. The question in this dimension can be selected as the next question for the iteration to improve accuracy in that dimension. To determine the most influential question, a degree can be determined for each input node of neural network 602 such that the change in the numerical rating input to the corresponding input node alters the probability curve (e.g., the degree to which the curve width changes) according to that degree. Based on this, the question associated with the input node whose degree of change in the probability curve is determined to be the highest can be selected as the most influential question for the corresponding iteration.

[0050] The high-level architecture of neural network 602 can also be advantageous because it allows for the easy integration of feedback into the neural network. As mentioned above, if the feedback represents a new input value that has not yet been input into neural network 602, a new input node can be simply added to neural network 602, and the new input value can be assigned to the new input node when training neural network 602. In this way, any type of new feedback can be easily integrated into the network, allowing neural network 602 to improve its ability to compute personality data. As an implementation method to reduce computational complexity when adding new input nodes, it is conceivable that when the network is trained to associate new input nodes with other nodes in the network, only those nodes determined to have the greatest impact relative to the best achievable result can be merged into the computation, thereby avoiding merging all nodes into the computation. Moreover, it is conceivable that when the network is trained to associate new input nodes with other nodes in the network, the number of pre-computed layers is limited (e.g., limited to 2 or 3) to, for example, avoid computing all subsequent node combinations.

[0051] In the above description, the techniques presented for efficiently retrieving digital representations of a user's personality data have been illustrated in the context of adapting a vehicle's driving configuration, such as adapting the vehicle's throttle and braking response behavior to a user's personality. In this context, the methods described herein can also be represented as methods for adapting a vehicle's driving configuration, including efficiently retrieving digital representations of a user's personality data. It should be understood that adapting a vehicle's throttle and braking response behavior is merely one example of adapting a vehicle's driving configuration, and more generally, adapting a vehicle's driving configuration can include any vehicle configuration that influences the vehicle's driving behavior. Thus, adapting a vehicle's driving configuration can include adapting at least one of the vehicle's throttle and braking response behavior, the vehicle's chassis settings, the vehicle's driving mode, and the vehicle's adaptive cruise control (ACC) settings to the user's personality. The vehicle's driving mode can include setting an economy mode, comfort mode, or sport mode to influence the vehicle's accelerator pedal and fuel consumption behavior based on the driver's personality. For example, if personality data indicates that the driver tends to be risk-averse, the driving mode can be set to economy or comfort mode, while for a driver with a risk-seeking personality, the driving mode can be set to sport mode. For example, the driving modes adapted to the vehicle may also include enabling / disabling the vehicle's automatic four-wheel drive (4WD) mode. ACC settings may include setting the distance to the vehicle ahead and / or the target driving speed, for example, depending on the driver's risk aversion.

[0052] It should be understood that the techniques presented herein can also be used for other purposes in vehicle scenarios, such as adapting the environmental conditions in the passenger cabin of a vehicle (or more generally, the environmental conditions in the passenger cabin of a transportation device, since adapting the environmental conditions in the driver's cab can be similarly applied to other transportation devices, such as airplanes, trains, etc.). In this case, the methods described herein can also be represented as methods for adapting the environmental conditions in the passenger cabin of a transportation device, including efficiently retrieving a digital representation of a user's personality data. Adapting the environmental conditions in the passenger cabin of a transportation device may include adapting at least one of the following, such as the cabin temperature (e.g., by adapting the cabin's air conditioning settings), the cabin's interior lighting, and the oxygen level in the cabin, to the user's personality. Additionally or alternatively, in order to adapt the environmental conditions in the passenger cabin, the techniques presented herein can also be used to adapt user-specific settings regarding the passenger cabin. Adapting user-specific settings regarding the passenger cabin of a transportation device may include adapting at least one of the following, such as the passenger seat configuration for the user (e.g., seat height, seat position, seat massage settings, seat belt tension, etc.) and the equalizer settings of the sound system provided to the user in the passenger cabin (e.g., increasing / decreasing bass or height), to the user's personality.

[0053] In addition to adapting to the user's personality, any of the above adaptations of the vehicle / transportation device settings may be performed by considering (or "based on") sensor data indicating the user's attention level obtained in the cabin. In other words, the client device can be configured to consider not only a digital representation of the user's personality data but also sensor data indicating the user's attention level to adapt to at least one of the vehicle's driving configuration, the environmental conditions in the cabin, and user-specific settings regarding the cabin. In other words, the digital representation of the user's personality data and the sensor data indicating the user's attention level can be combined before performing the aforementioned adaptations. Sensor data indicating the user's attention level may include, for example, data regarding at least one of the user's heart rate, respiration, fatigue level, reaction time, and alcohol / drug levels. For example, the sensor data may be collected by at least one sensor installed in the cabin or in the user's mobile terminal.

[0054] Figure 7An exemplary implementation is illustrated, which includes incorporating the driver's personality data to consider the driver's attention level in order to adapt the vehicle's driving configuration, cabin environmental conditions, and / or user-specific settings regarding the cabin. The driver's attention level can be checked by corresponding sensors based on, for example, the user's reaction time, fatigue, heart rate, respiration, alcohol / drug levels, or abnormal user behavior. In the left portion of the figure, the collected sensor data indicates the user's normal attention level, and therefore, vehicle settings can be maintained at a normal level (e.g., adapted to the driver's personality or "MindDNA"), including, for example, speed, volume, temperature, and seat settings. In the middle portion of the figure, the sensor data indicates a reduced driver attention level, and therefore, vehicle settings can be changed to a lower speed, higher volume, lower temperature setting, including activating the seat massage function to restore the driver's attention. Optionally, attention tests can be performed, such as requesting the driver to provide a voice-based response in a question / answer scheme, and the results of the attention test can be considered when adapting the settings mentioned above. On the other hand, in the right part of the figure, sensor data indicates a very low level of driver attention and can therefore provide a warning to the user, and can be adapted accordingly to vehicle settings, such as adapting the vehicle settings to very low speeds (and for example, forcibly stopping the vehicle at the next parking opportunity), adapting to silent audio, and / or adapting to, for example, providing directions to the next hotel via a navigation system.

[0055] To provide user-tailored services, as described above (e.g., by adapting to at least one of the vehicle's driving configuration, cabin environmental conditions, and user-specific settings regarding the cabin), the client device may also consider body scan data indicating user (e.g., physiological) characteristics deriveable from scanning the user's body (e.g., at least a portion thereof) before providing user-tailored services (e.g., before the user drives the vehicle). User characteristics deriveable from scanning the user's body may include at least one of, for example, the user's size, weight, gender, age, build, posture, and emotional state. Body scan data can be obtained via a camera or voice recorder (e.g., of the user's mobile terminal, or installed on a vehicle / transportation device) that acquires one or more images or voice signals of the user, wherein body / face / voice recognition technology can be used to scan the user's body and derive the aforementioned user characteristics. The client device can therefore be configured to provide user-tailored services considering not only a digital representation of the user's personality data but also (or "based on" / "according to") body scan data. In other words, the digital representation of the user's personality data and body scan data can be combined before providing user-tailored services to the user. Figure 8An exemplary implementation is illustrated, which involves incorporating driver personality data into consideration driver body scan data (e.g., body scan data obtained by the driver's mobile terminal (such as the driver's smartphone, smartwatch, or fitness tracker) before entering the vehicle) to adapt accordingly to the vehicle's driving configuration, the environmental conditions in the cabin, and / or user-specific settings regarding the cabin. In the figures, the body scan data is represented as "BodyDNA," which, combined with "MindDNA," forms what is called "LifeDNA." It should be understood that the obtained body scan data can also be used to provide feedback characterizing the user's response to updating the neural network, as described above.

[0056] In another vehicle-related use case, the techniques presented herein can also be used to determine a vehicle configuration suited to a user's personality prior to vehicle manufacturing, wherein the vehicle can then be manufactured based on (or "according to") the determined vehicle configuration. Vehicles can be manufactured with different configuration options (e.g., provided by the vehicle manufacturer), such as different motor options, each with different motor power, driving technology options (e.g., support for two-wheel drive (2WD) or 4WD technology), chassis options, different drive mode options, support for ACC, etc., and when a new vehicle is to be manufactured for a user, the vehicle configuration can be determined to be particularly suited to the user's personality. For example, if personality data indicates that a user tends to be risk-averse, the determined vehicle configuration could include selecting a motor with lower power compared to a vehicle configuration determined for a user whose personality data indicates a risk-seeking personality. Based on the determined vehicle configuration, the vehicle can then be manufactured accordingly. Thus, based on the above description, a method for vehicle manufacturing can also be envisioned, including a digital representation of a user's personality data efficiently retrieved from a server by a client device, the digital representation of the personality data being processed at the client device to provide a vehicle configuration suited to the user's personality. The method may include: sending a request from a client device to a server for a digital representation of a user's personality data; receiving the requested digital representation of the user's personality data from the server by the client device, the user's personality data being calculated based on input obtained from the user using a neural network trained to compute the user's personality data based on the input obtained from the user; processing the digital representation of the personality data to determine a vehicle configuration suitable for the user's personality; and manufacturing the vehicle based on the determined vehicle configuration. During the vehicle manufacturing process, it should be understood that the determined vehicle configuration may also affect the manufacturing of vehicle parts required to manufacture the vehicle. For example, manufacturing the vehicle may include manufacturing one or more vehicle components for manufacturing the vehicle, wherein the vehicle components are manufactured according to the determined vehicle configuration (e.g., using a 3D printer).

[0057] It should be understood that the techniques presented herein can be used not only for vehicle / transportation device related use cases, but also for other use cases, such as adapting the configuration of smart home appliances or robots to a user's personality. Therefore, based on the above description, a method for adapting the configuration of smart home appliances (e.g., automatic blinds, air conditioners, refrigerators, washing machines, televisions, set-top boxes, etc.) can also be envisioned, including efficiently retrieving a digital representation of the user's personality data, wherein the digital representation of the user's personality can be processed at a client device to adapt the configuration of the smart home appliance to the user's personality (e.g., to adapt the smart home appliance to perform its primary tasks (such as its closing (blinds), heating / cooling (air conditioners), cooling (refrigerators), washing (washing machines), or recording / displaying (television / set-top box) tasks)). Similarly, based on the above description, a method for adapting the configuration of robots (e.g., humanoid robots or home robots configured to perform one or more household tasks) can be envisioned, including efficiently retrieving a digital representation of the user's personality data, wherein the digital representation of the user's personality can be processed at a client device to adapt the robot's configuration to the user's personality (e.g., to adapt the home robot to perform household tasks).

[0058] Various other use cases are generally conceivable. For example, other use cases may include the adaptation of a virtual robot's configuration, the adaptation of a medical device's configuration, or even brain stimulation. Therefore, based on the above description, a method for adapting the configuration of a virtual robot (e.g., a chatbot, virtual service provider, or virtual personal assistant) can also be conceivable, including the efficient retrieval of a digital representation of a user's personality data, wherein the digital representation of the user's personality can be processed at a client device to adapt the virtual robot's configuration to the user's personality (e.g., to adapt the virtual robot to perform its tasks supporting the user). Similarly, based on the above description, a method for adapting the configuration of a medical device (e.g., a bedside medical device) can be conceivable, including the efficient retrieval of a digital representation of a user's personality data, wherein the digital representation of the user's personality can be processed at a client device to adapt the medical device's configuration to the user's personality (e.g., to adapt a dosing regimen, such as the administration of analgesics). Furthermore, a method for stimulating the brain (e.g., a biological brain or a virtual representation of the brain) can be envisioned, comprising the efficient retrieval of a digital representation of a user's personality data, wherein the digital representation of the user's personality can be processed at a client device to adapt the brain stimulation program based on the user's personality. The stimulation program may include, for example, electrical stimulation of a biological brain or adaptation / reconfiguration of a virtual representation of the brain. For instance, the virtual representation of the brain can be fed into a robot or other form of intelligent system to influence the behavior of such a system based on the user's personality.

[0059] In all the examples and use cases described above, when it is mentioned that a configuration or setting is “adapted to the user’s personality,” it should be understood that such adaptation can be achieved using a predefined mapping that maps a given characteristic of the user’s personality (as indicated by a numerical representation of the user’s personality data) to a specific configuration or setting of a corresponding device / apparatus (e.g., a vehicle, transportation vehicle, smart home appliance, robot, medical device, etc., as described above). For example, if personality data indicates that a driver tends to be risk-averse, the vehicle’s driving mode can be set to Eco or Comfort mode, while for a driver tending to have a risk-seeking personality, the driving mode can be set to Sport mode. Such mappings can be predefined for each possible combination of personality characteristic and configuration / setting, and the configuration or setting of the device / apparatus can be adapted accordingly based on the obtained user’s personality data. For example, as described above, a user’s personality characteristic can correspond to a value of a personality dimension (e.g., a personality dimension among the Big Five personality traits) output by a neural network.

[0060] It will be fully appreciated from the foregoing description that the advantages of the technology presented herein will be clear, and it will be apparent that various changes can be made in terms of form, construction, and arrangement of its exemplary aspects without departing from the scope of this disclosure or sacrificing all its advantageous effects. Because the technology presented herein can be varied in many ways, it will be understood that this disclosure should be limited only by the scope of the appended claims.

Claims

1. A method of retrieving, by a client device (502; 406), a digital representation of personality data of a user (402) from a server (404), the method being performed by the server (404) and comprising: storing (S202) a neural network (602) trained to compute personality data of a user (402) based on input obtained from the user (402); receiving (S204), from the client device (502; 406), a request for a digital representation of personality data of a user (402); and sending (S206), to the client device (502; 406), a digital representation of the requested personality data of the user (402), the digital representation of the personality data of the user (402) being processed at the client device (502; 406) to provide a user-adapted service to the user (402), wherein the personality data of the user (402) is computed using the neural network (602) based on input obtained from the user (402), and wherein the method further comprises: receiving feedback characterizing the user (402); updating the neural network (602) based on the feedback, wherein updating the neural network (602) comprises training the neural network (602) based on the feedback; and sending, to the client device (502; 406), a digital representation of updated personality data of the user (402), wherein the updated personality data of the user (402) is computed using the updated neural network (602).

2. The method of claim 1, wherein the digital representation of the personality data of the user (402) is processed at the client device (502; 406) to adapt a configuration of at least one device (406) providing a service to the user (402).

3. The method of claim 2, wherein, Adapting the configuration of the at least one device (406) to accommodate the personality of the user (402) is achieved using a mapping that maps characteristics of the personality of the user (402) indicated by the digital representation of the personality data of the user (402) to specific configurations of the at least one device (406).

4. The method of claim 2, wherein the at least one device (406) comprises the client device (406).

5. The method of claim 2, wherein the digital representation of the updated personality data of the user (402) is processed at the client device (502; 406) to improve the configuration of the at least one device (406) providing the service to the user (402).

6. The method of claim 2, wherein the feedback comprises behavior data reflecting behavior of the user (402) monitored at the at least one device (406) when using the service provided by the at least one device (406).

7. The method of claim 6, wherein, monitoring the behavioral data using measurements performed by the at least one device (406) providing the service to the user (402).

8. The method according to claim 6, wherein the at least one device (406) comprises a vehicle, and wherein the behavioral data comprises data reflecting driving behavior of the user (402).

9. The method according to any one of claims 1 to 8, wherein the feedback is indicative of a personality of the user (402).

10. The method according to any one of claims 1 to 8, wherein the personality data of the user (402) is indicative of at least one of: a psychological characteristic of the user (402), and a preference of the user (402).

11. The method according to any one of claims 1 to 8, wherein the input obtained from the user corresponds to a numerical score of answers to questions reflecting at least one of personality, goals and motivations of the user (402), and wherein when calculating the personality data of the user (402) using the neural network (602), each numerical score is used as input to a separate input node of the neural network (602).

12. The method according to claim 11, wherein the questions about the personality of the user (402) correspond to at least one of: the International Personality Item Pool, IPIP, the HEXACO-60 pool, the Big Five Personality Inventory-10 pool, i.e. the BFI-10 pool, questions about a psychological characteristic of the user (402), and questions about a preference of the user (402).

13. The method according to claim 11, wherein the questions correspond to questions selected from a set of questions representing a best achievable result of calculating personality data of a user (402), wherein the selected questions correspond to questions of the set of questions determined to be most influential with respect to the best achievable result.

14. The method according to claim 13, wherein the number of selected questions is less than 10% of the number of questions contained in the set of questions.

15. The method according to claim 13, wherein the questions are selected from the set of questions based on correlating an achievable result of each individual question of the set of questions with the best achievable result, and selecting questions from the set of questions having a highest correlation with the best achievable result.

16. The method of claim 13, wherein the questions are iteratively selected from the set of questions, wherein, in each iteration, the next question is selected according to the user’s answer to a previous question, and wherein, in each iteration, the next question is selected as a question of the set of questions determined to be most influential on an achievable result for calculating personality data of the user.

17. The method of claim 16, wherein the neural network (602) comprises a plurality of output nodes representing probability curves (604) of outcomes of the personality data of the user (402), wherein determining the most influential question of the set of questions as the next question of the respective iteration comprises: for each input node of the neural network (602), a degree is determined according to which a change in a numerical score input to the respective input node of the neural network (602) changes the probability curve (604).

16. The method according to any one of claims 1 to 15, wherein the neural network (602) is trained using a plurality of users (402) and a plurality of sets of questions, wherein each set of questions corresponds to a respective one of the plurality of users (402), and wherein each set of questions is selected based on a respective one of the plurality of users (402) answering the questions of the respective set of questions.

17. The method according to claim 16, wherein the plurality of sets of questions are selected based on a respective one of the plurality of users (402) answering the questions of the respective set of questions.

18. The method according to claim 16 or 17, wherein the plurality of sets of questions are selected based on a respective one of the plurality of users (402) answering the questions of the respective set of questions.

19. The method according to any one of claims 16 to 18, wherein the plurality of sets of questions are selected based on a respective one of the plurality of users (402) answering the questions of the respective set of questions.

20. The method according to any one of claims 16 to 19, wherein the plurality of sets of questions are selected based on a respective one of the plurality of users (402) answering the questions of the respective set of questions.

21. The method according to any one of claims 16 to 20, wherein the plurality of sets of questions are selected based on a respective one of the plurality of users (402) answering the questions of the respective set of questions.

22. The method according to any one of claims 16 to 21, wherein the plurality of sets of questions are selected based on a respective one of the plurality of users (402) answering the questions of the respective set of questions.

23. The method according to any one of claims 16 to 22, wherein the plurality of sets of questions are selected based on a respective one of the plurality of users (402) answering the questions of the respective set of questions.

24. The method according to any one of claims 16 to 23, wherein the plurality of sets of questions are selected based on a respective one of the plurality of users (402) answering the questions of the respective set of questions.

18. The method according to any one of claims 1 to 8, wherein the personality data of the user (402) is computed prior to receiving the request from the client device (502; 406), and wherein the request comprises an access code previously provided to the user (402) by the server (404) when computing the personality data of the user (402), the access code allowing the user (402) to access the digital representation of the personality data of the user (402) from a different client device (502; 406).

19. The method according to any one of claims 1 to 8, wherein providing the user (402) with the user-adapted service comprises stimulating a brain, wherein the digital representation of the personality data of the user (402) is processed at the client device (502) to adapt a stimulation program of a brain based on the personality of the user (402).

20. The method according to any one of claims 2 to 8, wherein providing the user (402) with the user-adapted service comprises one of: - adapting a driving configuration of a vehicle, wherein the at least one device comprises a vehicle (406), and wherein the digital representation of the personality data of the user (402) is processed at the client device (502; 406) to adapt a driving configuration of the vehicle (406) to the personality of the user (402), - adapting an environmental condition in a passenger cabin of a transportation means (406), wherein the at least one device comprises a transportation means (406), and wherein the digital representation of the personality data of the user (402) is processed at the client device (502; 406) to adapt an environmental condition in a passenger cabin of the transportation means (406) to the personality of the user (402), - adapting a user-specific setting with respect to a passenger cabin of a transportation means (406), wherein the at least one device comprises a transportation means (406), and wherein the digital representation of the personality data of the user (402) is processed at the client device (502; 406) to adapt a user-specific setting with respect to a passenger cabin of the transportation means (406) to the personality of the user (402), - adapting a configuration of a smart home appliance, wherein the at least one device comprises a smart home appliance, and wherein the digital representation of the personality data of the user (402) is processed at the client device (502) to adapt a configuration of the smart home appliance to the personality of the user (402), - adapting a configuration of a robot, wherein the at least one device comprises a robot, and wherein the digital representation of the personality data of the user (402) is processed at the client device (502) to adapt a configuration of the robot to the personality of the user (402), - adapting a configuration of a virtual robot, wherein the at least one device executes a virtual robot, and wherein the digital representation of the personality data of the user (402) is processed at the client device (502) to adapt the configuration of the virtual robot to the personality of the user (402), and - adapting a configuration of a medical device, wherein the at least one device comprises a medical device, and wherein the digital representation of the personality data of the user (402) is processed at the client device (502) to adapt the configuration of the medical device to the personality of the user (402).

21. The method of claim 20, wherein, When the at least one device (406) comprises a transportation means, it is also considered to perform providing the user-adapted service to the user (402) taking into account sensor data indicative of a level of attention of the user (402) obtained in a passenger cabin of the transportation means (406).

22. The method of claim 1, wherein providing the user-adapted service to the user comprises providing a vehicle configuration adapted to the personality of the user, wherein the digital of the personality data of the user is processed at a client device (110) to determine a vehicle configuration adapted to the personality of the user.

23. The method of claim 22, wherein a vehicle is manufactured based on the determined vehicle configuration.

24. The method of claim 22 or 23, wherein the digital representation of the updated personality data of the user is processed at the client device (100) to improve the vehicle configuration.

25. The method of any one of claims 1 to 8, wherein the feedback is collected at the client device (110).

26. The method of any one of claims 1 to 8, wherein, It is also considered to perform providing the user-adapted service to the user (402) taking into account body scan data indicative of a characteristic of the user (402) derivable by scanning at least one part of the body of the user (402).

27. A method for retrieving a digital representation of personality data of a user (402) from a server (404) via a client device (502; 406), the method being performed by the client device (502; 406) performing and comprising: sending (S302), to the server (404), a request for a digital representation of personality data of a user (402); receiving (S304), from the server (404), the requested digital representation of the personality data of the user (402), the personality data of the user (402) being computed based on input obtained from the user (402), using a neural network (602) trained to compute personality data of a user (402) based on input obtained from the user (402); and processing (S306) the digital representation of the personality data to provide a user-adapted service to the user (402), wherein the method further comprises: sending feedback characterizing the user (402) to the server (404); and receiving, from the server (404), a digital representation of updated personality data of the user (402), wherein the updated personality data of the user (402) is computed using the neural network (602) that is updated based on the feedback, wherein updating the neural network comprises training the neural network based on the feedback.

28. The method of claim 27, wherein the digital representation of the personality data of the user (402) is processed at the client device (502; 406) to adapt a configuration of at least one device (406) providing a service to the user (402).

29. The method of claim 28, wherein, Adapting the configuration of the at least one device (406) to accommodate the personality of the user (402) is achieved using a mapping that maps characteristics of the personality of the user (402) indicated by the digital representation of the personality data of the user (402) to specific configurations of the at least one device (406).

30. The method of claim 28, wherein the at least one device (406) comprises the client device (406).

31. The method of claim 28, wherein the digital representation of updated personality data of the user (402) is processed at the client device (502; 406) to improve the configuration of the at least one device (406) providing a service to the user (402).

32. The method of claim 28, wherein the feedback comprises behavior data reflecting behavior of the user (402) monitored at the at least one device (406) when using a service provided by the at least one device (406).

33. The method of claim 32, wherein, The behavior data is monitored using measurements performed by the at least one device (406) providing the service to the user (402).

34. The method of claim 32, wherein the at least one device (406) comprises a vehicle, and wherein the behavior data comprises data reflecting driving behavior of the user (402).

35. The method of any one of claims 27 to 34, wherein the feedback indicates a personality of the user (402).

36. The method of any one of claims 27 to 34, wherein the personality data of the user (402) indicates at least one of: a psychological characteristic of the user (402), and a preference of the user (402).

37. The method of any one of claims 27 to 34, wherein the input obtained from the user corresponds to a numerical score of answers to questions reflecting at least one of personality, purpose, and motivation with respect to the user (402), and wherein when computing the personality data of the user (402) using the neural network (602), each numerical score is used as an input to a separate input node of the neural network (602).

38. The method according to claim 37, wherein the questions about the personality of the user (402) correspond to questions from at least one of the following: the International Personality Item Pool, IPIP, the HEXACO-60 pool, the Big Five Inventory-10 pool, i.e. the BFI-10 pool, questions about psychological characteristics of the user (402), and questions about preferences of the user (402).

39. The method according to claim 37, wherein the questions correspond to questions selected from a set of questions representing a best achievable outcome of computing personality data of a user (402), wherein the selected questions correspond to questions of the set of questions determined to be most influential with respect to the best achievable outcome.

40. The method according to claim 39, wherein the number of selected questions is less than 10% of the number of questions contained in the set of questions.

41. The method according to claim 39, wherein the questions are selected from the set of questions based on correlating achievable outcomes of each individual question of the set of questions with the best achievable outcome and selecting questions from the set of questions having the highest correlation with the best achievable outcome.

42. The method of claim 39, wherein the questions are iteratively selected from the set of questions, wherein, In each iteration, the next question is selected according to the user’s answer to the previous question, and wherein in each iteration, the next question is selected as the question of the set of questions determined to be most influential on the achievable outcome for computing personality data of the user.

43. The method of claim 42, wherein the neural network (602) comprises a plurality of output nodes representing probability curves (604) of outcomes of the personality data of the user (402), wherein determining the most influential question of the set of questions as the next question of the respective iteration comprises: for each input node of the neural network (602) a degree according to which a change in the numerical score aspect input to the respective input node of the neural network (602) changes the probability curve (604).

44. The method according to any one of claims 27 to 34, wherein the personality data of the user (402) is computed prior to receiving the request from the client device (502; 406), and wherein the request comprises an access code previously provided to the user (402) by the server (404) when computing the personality data of the user (402), the access code allowing the user (402) to access the digital representation of the personality data of the user (402) from a different client device (502; 406).

45. The method according to any one of claims 27 to 34, wherein providing the user (402) with the user-adapted service comprises stimulating the brain, wherein the digital representation of the personality data of the user (402) is processed at the client device (502) to adapt a stimulation program of the brain based on the personality of the user (402).

46. The method according to any one of claims 28 to 34, wherein providing the user (402) with the user-adapted service comprises one of: - adapting a driving configuration of a vehicle, wherein the at least one device comprises a vehicle (406), and wherein the digital representation of the personality data of the user (402) is processed at the client device (502; 406) to adapt a driving configuration of the vehicle (406) to the personality of the user (402), - adapting an environmental condition in a passenger cabin of a transportation means (406), wherein the at least one device comprises a transportation means (406), and wherein the digital representation of the personality data of the user (402) is processed at the client device (502; 406) to adapt an environmental condition in a passenger cabin of the transportation means (406) to the personality of the user (402), - adapting a user-specific setting with respect to a passenger cabin of a transportation means (406), wherein the at least one device comprises a transportation means (406), and wherein the digital representation of the personality data of the user (402) is processed at the client device (502; 406) to adapt a user-specific setting with respect to a passenger cabin of the transportation means (406) to the personality of the user (402), - adapting a configuration of a smart home appliance, wherein the at least one device comprises a smart home appliance, and wherein the digital representation of the personality data of the user (402) is processed at the client device (502) to adapt a configuration of the smart home appliance to the personality of the user (402), - adapting a configuration of a robot, wherein the at least one device comprises a robot, and wherein the digital representation of the personality data of the user (402) is processed at the client device (502) to adapt a configuration of the robot to the personality of the user (402), - adapting a configuration of a virtual robot, wherein the at least one device executes a virtual robot, and wherein the digital representation of the personality data of the user (402) is processed at the client device (502) to adapt a configuration of the virtual robot to the personality of the user (402), and - adapting a configuration of a medical device, wherein the at least one device comprises a medical device, and wherein the digital representation of the personality data of the user (402) is processed at the client device (502) to adapt a configuration of the medical device to the personality of the user (402).

47. The method of claim 46, wherein, When the at least one device (406) comprises a transportation means, sensor data indicative of an attention level of the user (402) obtained in a passenger cabin of the transportation means (406) is also considered to perform providing the user-adapted service to the user (402).

48. The method of claim 27, wherein providing the user-adapted service to the user comprises providing a vehicle configuration adapted to the personality of the user, wherein the digital of the personality data of the user is processed at a client device (110) to determine a vehicle configuration adapted to the personality of the user.

49. The method of claim 48, wherein a vehicle is manufactured based on the determined vehicle configuration.

50. The method of claim 48 or 49, wherein the digital representation of the updated personality data of the user is processed at the client device (100) to improve the vehicle configuration.

51. The method of any one of claims 27 to 34, wherein the feedback is collected at the client device (110).

52. The method of any one of claims 27 to 34, wherein, It is also considered to perform providing the user (402) with the user-adapted service by indicating body scan data that is derivable by scanning at least a part of the body of the user (402) for the characteristics of the user (402).

53. A computer program product comprising program code portions for performing the method according to any one of claims 1 to 52 when the computer program product is executed on one or more computing units.

54. The computer program product according to claim 53, stored on one or more computer-readable recording media.

55. A server (100; 404) enabling a client device (502; 406) to retrieve a digital representation of personality data of a user (402) from the server (404), the digital representation of personality data being processed at the client device (502; 406) to provide the user (402) with a user-adapted service, the server (404) comprising at least one processor (102) and at least one memory (104), the at least one memory (104) containing instructions executable by the at least one processor (102) to enable the server (404) to operate to perform the method according to any one of claims 1 to 26.

56. A client device (110; 502; 406) enabled to retrieve a digital representation of personality data of a user (402) from a server (404), the client device (110; 502; 406) comprising at least one processor (112) and at least one memory (114), the at least one memory (114) containing instructions executable by the at least one processor (112) to enable the client device (110; 502; 406) to operate to perform the method according to any one of claims 27 to 52.

57. An electronic system comprising a server (100; 404) according to claim 55 and at least one client device (110; 502; 406) according to claim 56.

58. A method of retrieving a digital representation of personality data of a user, the digital representation of personality data being processed to provide the user with a user-adapted service, the method comprising: obtaining a digital representation of personality data of a user, the personality data of the user being computed based on input obtained from the user using a neural network trained to compute personality data of a user based on input obtained from the user; and processing the digital representation of the personality data to provide a user-adapted service to the user, wherein the method further comprises: obtaining feedback characterizing the user; and obtaining an updated digital representation of personality data of the user (402), wherein the updated personality data of the user (402) is computed using the neural network (602) updated based on the feedback, wherein updating the neural network comprises training the neural network based on the feedback.

Images