A system and method for payment management of electric vehicles

Abstract

The present invention provides a system (100) for payment management of electric vehicles. The system (100) comprises plurality of modules including a data collection module (103), a data pre-processor module (104), a data analysis module (105), a feature extraction module (106), and a prediction module (107). The data collection module (103) is configured to receive a set of data. The data pre-processor module (104) is configured to conduct one or more operations on the set of data for obtaining a pre-processed data. The data analysis module (105) is configured to perform one or more statistical analysis operations on the pre-processed data to obtain a feature set. The feature extraction module (106) is configured to select suitable features from the feature set. The prediction module (107) is configured to process the selected feature set for predicting the ability of a user to pay the corresponding due amount.

Description

[0001] “A SYSTEM AND METHOD FOR PAYMENT MANAGEMENT OF ELECTRIC VEHICLES”

[0002] FIELD OF THE INVENTION

[0003] The present invention relates to the field for payment management in electric vehicles. More particularly, the present invention relates to a system and method for payment management of electric vehicles.

[0004] BACKGROUND OF THE INVENTION

[0005] With the rising adoption of electric vehicles across the globe, their appeal is driven by environmental advantages, lower operating costs, and the push for sustainable energy solutions. Continuous advancements in battery technology have extended ranges, while the rapid expansion of charging infrastructure has eased concerns over convenience and accessibility. Coupled with government incentives and stricter emission regulations, EVs are becoming a mainstream choice for both consumers and manufacturers. This sift is not only transforming the automotive sector but also driving the development of cleaner, more energy-efficient transportation systems globally.

[0006] As the EV market is expanding, there is a growing need to streamline the billing and payment process for both users and service providers. Managing billing, payments, and operational logistics is crucial in the evolving landscape of EV adoption and charging infrastructure. Service providers face the significant task of efficiently handling billing processes related to electricity usage, which includes monitoring and invoicing for charging services. Moreover, they must ensure the timely collection of payments from users, encompassing Equated Monthly Installments (EMIs) for financed EV purchases and regular payments for ongoing charging services. This operational framework is essential for sustaining reliable service delivery and financial stability amid the increasing complexity of EV infrastructure management. Conventional methods of billing and payment collection in the EV charging Industry often involve manual processes, such as generating invoices and sending them to users at regular intervals. This reliance on human intervention introduce human errors and inconsistencies, potentially leading to delays in payment processing and revenue loss for service providers. Additionally, standard payment reminders, typically sent through conventional channels like email or postal email, may not always reach users promptly or effectively, further exacerbating the issue of delayed payments. For instance, if a service provider sends out monthly invoices via postal mail, users may not receive them on time due to postal delays or may misplace them altogether. This situation can result in missed payments and subsequent late fees, negatively impacting both user satisfaction and the service provider’s cash flow. Moreover, the complexity of tracking electricity consumption and payment schedules adds additional layer of challenge. Users may find it cumbersome to manually monitor their usage patterns against billing cycles, especially in dynamic charging environments where usage can vary widely. This lack of real-time visibility into consumption data can lead to discrepancies between expected and actual bills, causing confusion for both users and service provides alike.

[0007] These challenges are compounded by the operational intensity required to monitor and manage payments effectively. Service providers frequently encounter difficulties in predicting payment behaviors and implementing efficient collection strategies. Tike, the need for consistent tracking of EMIs and timely reminders to users is important to prevent payment delays or defaults, yet this can create a significant administrative burden.

[0008] US9505317B2 discloses system and method for electric vehicle charging and billing using a wireless vehicle communication service. However, this invention provides the billing facility, however does not have the provision manage the payment and this invention does not have the ability to predict the user’s ability to pay the amount.

[0009] US20240140251A1 discloses systems, devices, and methods relate to managing vehicle battery charging. Payment for charging is preauthorized, and a receipt for preauthorization is sent from a payment management device to a user device when the user device has access to internet or cellular communications. This preauthorization receipt is later sent from the user device to a charge station where internet or cellular communication is not available, to enable charging of the vehicle battery at the charge station. Another receipt is sent from the charge station to the user device after charging of the vehicle, indicating a payment amount for actual energy used to charge the battery. Later, when the user device has access to cellular or internet communications, the user device sends the receipt indicating payment amount for actual energy used to the payment management device, for final payment balancing. However, this invention does not have the provision to manage the payment and this invention does not have the ability to predict the user’s ability to pay the amount.

[0010] Therefore, there is a need of a system and method for user payment management of electric vehicles.

[0011] OBJECT OF THE INVENTION

[0012] The main object of the present invention is to provide a system and method for user payment management of electric vehicles.

[0013] Another object of the present invention is to provide a solution to achieve operational efficiency in payment and collection which reduce operating expenses or expenditure and mitigate delayed payments.

[0014] Yet another object of the present invention is to provide a solution that track parameters which directly correlates to the ability of the user and their impact to timely payments and propensity to pay.

[0015] Yet another object of the present invention is to provide a solution that utilizes advanced intelligent techniques for real time prediction of the ability of the user to pay the bill or equated monthly instalment (EMI).

[0016] Still another object of the present invention is to provide a solution which integrates multi-parameter tracking, advanced techniques, and a continuous feedback loop, and offers a more accurate and efficient method and system for predicting and managing payments and defaults.

[0017] SUMMARY OF THE INVENTION

[0018] The present invention relates to a system and method for payment management of electric vehicles by integrating features of multi-parameter tracking, advanced techniques, and continuous feedback loop.

[0019] In an embodiment, the present invention provides a system for payment management of electric vehicles. The system comprises a processor and data storage unit connected with the processor. The processor includes a plurality of modules including a data collection module, a data pre-processor module, a data analysis module, a feature extraction module, a prediction module. The data collection module is configured to receive a set of data from one or more sources. The set of data include but not limited to a battery data, a behavioral data. The battery data further include but not limited to location, daily running, and earnings battery health. The behavioral data include but not limited to age, social score, telco score, identity confidence, payment history, credit score, and loan amount. The data pre-processor module is configured to conduct one or more operations on the set of data to remove the unwanted data, thereby obtaining a pre-processed data. The data analysis module is configured to perform one or more statistical analysis operations on the pre- processed data to obtain a feature set. The feature extraction module is configured to select suitable features from the feature set which are most correlated for better prediction. The prediction module is configured to process, via one or more machine learning and / or artificial intelligence techniques, the selected feature set for predicting the ability of a user to pay the corresponding due amount.

[0020] In another embodiment, the present invention provides a method for payment management for electric vehicles. The method comprises steps of (a) receiving, via a data collection module, a set of data from one or more sources. The set of data include but not limited to a battery data, a behavioral data. The battery data further include but not limited to location, daily running, and earnings battery health. The behavioral data include but not limited to age, social score, telco score, identity confidence, payment history, credit score, and loan amount. Further, at step (b), conducting, via a data pre-processor module, one or more operations on the set of data to remove the unwanted data, thereby obtaining a pre-processed data. Further, at step (c), performing one or more statistical analysis operations on the pre-processed data to obtain a feature set. Further, at step (d), selecting, via a feature extraction module, suitable features from the feature set which are most correlated for better prediction. Further, at step (e), processing, via a prediction module by one or more machine learning and / or artificial intelligence techniques, the selected feature set for predicting the ability of a user to pay the corresponding due amount.

[0021] In another embodiment, the present invention provides a user equipment for payment management for electric vehicles. The user equipment module is configured to receive a set of data from one or more sources. The set of data include but not limited to a battery data, a behavioral data. The battery data further include but not limited to location, daily running, and earnings battery health. The behavioral data include but not limited to age, social score, telco score, identity confidence, payment history, credit score, and loan amount. The user equipment module is further configured to conduct one or more operations on the set of data to remove the unwanted data, thereby obtaining a pre-processed data. The user equipment module is further configured to perform one or more statistical analysis operations on the pre- processed data to obtain a feature set. The user equipment module is further configured to select suitable features from the feature set which are most correlated for better prediction. The user equipment module is further configured to process, via one or more machine learning and / or artificial intelligence techniques, the selected feature set for predicting the ability of a user to pay the corresponding due amount.

[0022] The above objects and advantages of the present invention will become apparent from the hereinafter set forth brief description of the drawings, detailed description of the invention, and claims appended herewith.

[0023] BRIEF DESCRIPTION OF THE DRAWINGS An understanding of the system and method for payment management of electric vehicles of the present invention may be obtained by reference to the following drawings:

[0024] Figure 1 is a block diagram of the system for payment management of electric vehicles according to an embodiment of the present invention.

[0025] Figure 2 is a flow chart of the method for payment management of electric vehicles according to an embodiment of the present invention.

[0026] Figure 3 is another flow chart of the method for payment management of electric vehicles according to a preferred embodiment of the present invention.

[0027] DETAILED DESCRIPTION OF THE INVENTION

[0028] The present invention will now be described hereinafter with reference to the accompanying drawings in which a preferred embodiment of the invention is shown. This invention may, however, be embodied in many different forms and should not be construed as being limited to the embodiment set forth herein. Rather, the embodiment is provided so that this disclosure will be thorough, and will fully convey the scope of the invention to those skilled in the art.

[0029] Many aspects of the invention can be better understood with references made to the drawings below. The components in the drawings are not necessarily drawn to scale. Instead, emphasis is placed upon clearly illustrating the components of the present invention. Moreover, like reference numerals designate corresponding parts through the several views in the drawings. Before explaining at least one embodiment of the invention, it is to be understood that the embodiments of the invention are not limited in their application to the details of construction and to the arrangement of the components set forth in the following description or illustrated in the drawings. The embodiments of the invention are capable of being practiced and carried out in various ways. In addition, the phraseology and terminology employed herein are for the purpose of description and should not be regarded as limiting. The present invention now will be described hereinafter with reference to the detailed description, in which some, but not all embodiments of the invention are indicated. Indeed, the invention may be embodied in many different forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that this disclosure will satisfy applicable legal requirements. Like numbers refer to like elements throughout. The present invention is described fully herein with non-limiting embodiments and exemplary experimentation.

[0030] The present invention relates to a system and method for payment management of electric vehicles by integrating features of multi-parameter tracking, advanced techniques, and a continuous feedback loop.

[0031] In an embodiment, the present invention provides a system for payment management of electric vehicles. The system comprises a processor and data storage unit connected with the processor. The processor include a plurality of modules including a data collection module, a data pre-processor module, a data analysis module, a feature extraction module, a prediction module. The data collection module is configured to receive a set of data from one or more sources. The set of data include but not limited to a battery data, a behavioral data. The battery data further include but not limited to location, daily running, and earnings battery health. The behavioral data include but not limited to age, social score, telco score, identity confidence, payment history, credit score, and loan amount. The data pre-processor module is configured to conduct one or more operations on the set of data to remove the unwanted data, thereby obtaining a pre-processed data. The data analysis module is configured to perform one or more statistical analysis operations on the pre-processed data to obtain a feature set. The feature extraction module is configured to select suitable features from the feature set which are most correlated for better prediction. The prediction module is configured to process, via one or more machine learning and / or artificial intelligence techniques, the selected feature set for predicting the ability of a user to pay the corresponding due amount.

[0032] Figure 1 is a block diagram of the system for payment management of electric vehicles according to an embodiment of the present invention. The system (100) comprises a processor ( 102) and data storage unit (101) connected with the processor. The processor includes a plurality of modules such as a data collection module (103), a data pre-processor module (104), a data analysis module (105), a feature extraction module (106), and a prediction module (107).

[0033] The data collection module (103) is configured to receive a set of data from one or more sources such as data repository, online database, and offline database. In an implementation, the data collection module (103) is configured to operate in online, offline and hybrid mode. For example, the data collection module (103) is able to automatically collect the data from a user profile or data storage unit (101). In another instance, an operator provides instructions to the data collection module (103) for collecting the set of data from the suitable source.

[0034] The set of data include but not limited to a battery data, a behavioral data, loan, EMI, operational and behavioral parameters, payment behaviour history, creditworthiness indicators. The battery data further include but not limited to location, daily running, and battery health. The behavioral data include but not limited to age, social score, telco score, identity confidence, payment history, credit score, and loan amount. The set of data represent the financial capacity, operational dependency, earning potential and historical repayment discipline of the user. The present invention does not limit the scope of the set of data and further may include other details also.

[0035] The data pre-processor module (104) is configured to conduct one or more operations on the set of data to remove the unwanted data, thereby obtaining a pre-processed data. The operations used herein refers to pre-processing operations such as data cleaning, data integration, data transformation, data reduction, data discretization, data normalization, more specifically, missing value handling, outlier treatment, normalization and scaling, feature encoding. The output from the data pre-processor module (104) is structured, machine learning ready dataset.

[0036] The data analysis module (105) is configured to perform one or more statistical analysis operations on the pre-processed data to obtain a feature set. The statistical analysis operations are based on machine learning and / or artificial intelligence techniques. The data analysis module (105) performs statistical and correlation analysis to understand how each parameter affects repayment behavior.

[0037] The feature extraction module (106) is configured to select suitable features from the feature set which are most correlated for better prediction. The feature extraction module (106) also utilizes the machine learning and / or artificial intelligence techniques for selecting the suitable features such as loan behaviour parameters, recent repayment patterns, operational activity, demographic stability and financial credibility. Further the suitable features are then passes as the input vector to the prediction module (107).

[0038] The prediction module (107) is configured to process, via one or more machine learning and / or artificial intelligence techniques, the selected feature set for predicting the ability of a user to pay the corresponding due amount. In an implementation, the prediction module (107) process a set of parameters in the selected feature set such as risk score (based on 45 parameters), social footprint score, identity confidence score, karma score (daily running km, service turnaround time (TAT) (100% uptime), NFS), digital footprint, loan tenure, down payment, residence status, vehicle ownership, earning behavior.

[0039] The prediction module (107) is configured to apply the machine learning technique to identify a hidden correlation and a behavioural pattern within the set of parameters which are considered an input vector by the prediction module. Thereafter, the prediction module (107) is configured to compute a payment propensity score based on the identified hidden correlation and the behavioral pattern representing a probability of the user making the payment of the corresponding due amount. Further, the prediction module (107) is configured to generate the prediction from the payment propensity score and thereafter comparing the prediction to an actual behaviour displayed by the user. Furthermore, the prediction module (107) is configured to determine a deviation or tolerance between the prediction and the actual behaviour and presenting the deviation as a feedback input into the prediction module for updating a plurality of weights which are assigned to the set of parameters and thereby increasing an accuracy of the prediction. In other words, compute a payment propensity score for each user, classify the user as likely to pay on time or likely to pay late. The present invention compare predicted results with actual repayment behavior and apply a feedback loop to update the weights, increasing accuracy over time.

[0040] Figure 2 is a flow chart of the method for payment management of electric vehicles according to an embodiment of the present invention. In an implementation, the method is executed by the system (100).

[0041] The method comprises steps of (a) receiving, via the data collection module (103), a set of data from one or more sources. The set of data include but not limited to a battery data, a behavioral data. The battery data further include but not limited to location, daily running, and earnings battery health. The behavioral data include but not limited to age, social score, telco score, identity confidence, payment history, credit score, and loan amount.

[0042] Further, at step (b), conducting, via the data pre-processor module (104), one or more operations on the set of data to remove the unwanted data, thereby obtaining a pre- processed data.

[0043] Further, at step (c), performing, via the data analysis module, one or more statistical analysis operations on the pre-processed data to obtain a feature set.

[0044] Further, at step (d), selecting, via the feature extraction module (106), a set of parameters (i.e. suitable features) from the feature set which are most correlated for better prediction.

[0045] Further, at step (e), processing, via the prediction module (107) by one or more machine learning and / or artificial intelligence techniques, the selected feature set for predicting the ability of a user to pay the corresponding due amount.

[0046] Thereafter, the method further comprises the steps of applying, via the prediction module, the machine learning technique to identify a hidden correlation and a behavioral pattern within the set of parameters which are considered an input vector by the prediction module, computing, via the prediction module, a payment propensity score based on the identified hidden correlation and the behavioral pattern representing a probability of the user making the payment of the corresponding due amount, generating, via the prediction module, the prediction from the payment propensity score and thereafter comparing the prediction to an actual behaviour displayed by the user and determining, via the prediction module, a deviation or tolerance between the prediction and the actual behaviour and presenting the deviation as a feedback input into the prediction module for updating a plurality of weights which are assigned to the set of parameters and thereby increasing an accuracy of the prediction.

[0047] The present invention also provides a user equipment for payment management of electric vehicles. The user equipment module is configured to receive a set of data from one or more sources. The set of data include but not limited to a battery data, a behavioral data. The battery data further include but not limited to location, daily running, and earnings battery health. The behavioral data include but not limited to age, social score, telco score, identity confidence, payment history, credit score, and loan amount.

[0048] The user equipment module is further configured to conduct one or more operations on the set of data to remove the unwanted data, thereby obtaining a pre-processed data. The user equipment module is further configured to perform one or more statistical analysis operations on the pre-processed data to obtain a feature set. The user equipment module is further configured to select suitable features from the feature set which are most correlated for better prediction.

[0049] Thereafter, the user equipment module is configured to process, via one or more machine learning and / or artificial intelligence techniques, the selected feature set for predicting the ability of a user to pay the corresponding due amount.

[0050] EXAMPLE 1

[0051] Preferred Implementation of the present invention

[0052] The present invention provides the system (100) and method for payment management of electric vehicles that increases an operational efficiency in payment & collection, reduced operating expenses or expenditure and mitigate delayed payments & negative debts.

[0053] Figure 3 is another flow chart of the method for payment management of electric vehicles according to a preferred embodiment of the present invention. The system (100) and method of the present invention track parameters which directly correlate to the user ability and their impact to timely payments and propensity to pay. Based on which, the present invention identifies patterns & co-relations, build a model for real time prediction. Further, the prediction is then compared to actual behaviors displayed by the payer, match with the predicted outcome, compute the tolerance and feed in as an input to the model for improvement and increasing accuracy.

[0054] The present invention tracks both direct and indirect parameters influencing the payer’s ability and propensity to pay EMIs on time and includes a mechanism for continuous learning and adjustment of the prediction models based on actual behaviors observed. The present invention significantly reduces the time and resources required to monitor and manage payments and collections. Further, the predictive capabilities of the present invention help in proactive management, reducing instances of late payments and defaults.

[0055] EXAMPLE 2 Experimentation Analysis

[0056] The system and method for payment management of electric vehicles achieve an overall accuracy of 90 to 95%, and during testing on real payment data, the parameters considered are represented in below Table 1.

[0057] Table 1: Parameters and performance of the present invention

[0058] These results demonstrate that the present invention is highly effective in identifying users who will pay on time, leading to an overall predictive accuracy of -92%. The system further improves accuracy through a feedback learning mechanism, each wrong prediction updates the present invention, making future results more reliable. Also, the present invention operates in real time and generates the prediction within 2 to 3 seconds per user.

[0059] Therefore, the present invention provides system and method for payment management of electric vehicles, which integrates multi-parameter tracking, advanced techniques, and a continuous feedback loop, and offers a more accurate and efficient method and system for predicting and managing payments and defaults.

[0060] Many modifications and other embodiments of the invention set forth herein will readily occur to one skilled in the art to which the invention pertain having the benefit of the teachings presented in the foregoing descriptions and the associated drawings. Therefore, it is to be understood that the invention is not to be limited to the specific embodiments disclosed and that modifications and other embodiments are intended to be included within the scope of the appended claims. Although specific terms are employed herein, they are used in a generic and descriptive sense only and not for purposes of limitation.

[0061] The foregoing description of embodiments of the invention has been presented for purposes of illustration and description. It is not intended to be exhaustive or to limit the invention to the precise form disclosed, and modifications and variations are possible in light of the above teachings or may be acquired from practice of the invention. The embodiments were chosen and described in order to explain the principle of the invention and its practical application to enable one skilled in the art to utilize the invention in various embodiments and with various modifications as are suited to the particular use contemplated.

Claims

CLAIMS aim:

1. A system (100) for payment management of electric vehicles comprises: a processor (102); and a data storage unit (101) connected with the processor (102); wherein: said processor (102) includes a data collection module (103), a data preprocessor module (104), a data analysis module (105), a feature extraction module (106) and a prediction module (107); said data collection module (103) is configured to receive a set of data related to payment from one or more sources; said data pre-processor module (104) is configured to conduct one or more operations on the set of data to remove an unwanted data, thereby obtaining a pre-processed data; said data analysis module (105) is configured to perform a set of statistical analysis operations on the pre-processed data to obtain a feature set; said feature extraction module (106) is configured to select a set of parameters from the feature set which are correlated for prediction; said prediction module (107) is configured to process, via a machine learning technique, the set of parameters for predicting the ability of a user to pay a corresponding due amount.

2. The system (100) for payment management of electric vehicles as claimed in the claim 1 , wherein said one or more sources include data repository, online database and offline database.

3. The system (100) for payment management of electric vehicles as claimed in the claim 1 , wherein said data collection module ( 103) is configured to operate in online, offline and hybrid mode.

4. The system (100) for payment management of electric vehicles as claimed in the claim 1 , wherein said set of data includes battery data such as location, daily running and battery health, behavioural data such as age, social score, telco score, identity confidence, payment history, credit score and loan amount.

5. The system (100) for payment management of electric vehicles as claimed in the claim 1, wherein said one or more operations include data cleaning, data integration, data transformation, data reduction, data discretization, data normalization.

6. The system (100) for payment management of electric vehicles as claimed in the claim 1, wherein said set of parameters include social footprint score, identity confidence score, karma score (daily running km, service turnaround time (TAT) (100% uptime), net promoter score (NFS)), digital footprint, loan tenure, down payment, residence status, vehicle ownership and earning behaviour.

7. The system (100) for payment management of electric vehicle as claimed in claim 1, wherein said set of parameters are directly correlated to a financial ability of the user and likelihood of making payment of the corresponding amount.

8. The system (100) for payment management of electric vehicle as claimed in claim 1, wherein said prediction module (107) is configured to: a) apply the machine learning technique to identify a hidden correlation and a behavioural pattern within the set of parameters which are considered an input vector by the prediction module (107);b) compute a payment propensity score based on the identified hidden correlation and the behavioural pattern representing a probability of the user making the payment of the corresponding due amount; c) generate the prediction from the payment propensity score and thereafter comparing the prediction to an actual behaviour displayed by the user; d) determine a deviation or tolerance between the prediction and the actual behaviour and presenting the deviation as a feedback input into the prediction module (107) for updating a plurality of weights which are assigned to the set of parameters and thereby increasing an accuracy of the prediction.

9. The system (100) for payment management of electric vehicle as claimed in claim 1, wherein said system (100) operates in real time and predicts the ability of a user to pay a corresponding due amount within 2 to 3 seconds.

10. A method for payment management of electric vehicle comprises step of: a) receiving, via a data collection module (103), a set of data from one or more sources; b) conducting, via a data pre-processor module ( 104), one or more operations on the set of data obtained from step (a) to remove the unwanted data, thereby obtaining a pre-processed data; c) performing, via a data analysis module (105), a set of statistical analysis operations on the pre-processed data that is obtained from step (b) to further obtain a feature set; d) selecting, via a feature extraction module (106), a set of parameters from the feature set obtained from step (c) which are correlated for prediction; e) processing, via a prediction module (107) by a machine learning technique, the set of features for predicting the ability of a user to pay a corresponding due amount.

11. The method for payment management of electric vehicles as claimed in the claim 10, wherein said one or more sources include data repository, online database and offline database.

12. The method for payment management of electric vehicles as claimed in the claim 10, wherein said set of data includes battery data such as location, daily running and battery health, behavioural data such as age, social score, telco score, identity confidence, payment history, credit score and loan amount.

13. The method for payment management of electric vehicles as claimed in the claim 10, wherein said one or more operations include data cleaning, data integration, data transformation, data reduction, data discretization, data normalization.

14. The method for payment management of electric vehicles as claimed in the claim 10, wherein said set of parameters include social footprint score, identity confidence score, karma score (daily running km, service turnaround time (TAT) (100% uptime), net promoter score (NPS)), digital footprint, loan tenure, down payment, residence status, vehicle ownership, earning behaviour.

15. The method for payment management of electric vehicles as claimed in the claim 10, wherein said method further comprises the steps of: a) applying, via the prediction module (107), the machine learning technique to identify a hidden correlation and a behavioural pattern within the set of parameters which are considered an input vector by the prediction module (107); b) computing, via the prediction module (107), a payment propensity score based on the identified hidden correlation and the behavioural pattern representing a probability of the user making the payment of the corresponding due amount;c) generating, via the prediction module (107), the prediction from the payment propensity score and thereafter comparing the prediction to an actual behaviour displayed by the user; and d) determining, via the prediction module (107), a deviation or tolerance between the prediction and the actual behaviour and presenting the deviation as a feedback input into the prediction module (107) for updating a plurality of weights which are assigned to the set of parameters and thereby increasing an accuracy of the prediction.

Images