Method and apparatus for diagnosing automobile transmission

1 . A method comprising:

collecting vehicle information comprising transmission shift data of a vehicle;

pre-processing and labeling the transmission shift data;

among feature sets stored in a database, determining, as a plurality of candidate feature sets, a plurality of feature sets with respect to a plurality of other vehicles each associated with vehicle information that is same as at least a portion of the vehicle information of the vehicle;

generating a candidate feature set list comprising the plurality of candidate feature sets;

generating, using a supervised learning-based artificial intelligence model and for each of the plurality of candidate feature sets, an evaluation index of a transmission shift comfort level associated with the transmission shift data; and

outputting a feature set of the vehicle for diagnosing a transmission equipped in the vehicle by selecting, among candidate feature sets having evaluation indexes above a threshold value, a candidate feature set having a highest evaluation index.

2 . The method of claim 1 , further comprising:

based on none of a second plurality of candidate feature sets having an evaluation index above the threshold value, generating a new feature set based on additional transmission shift data associated with the vehicle.

3 . The method of claim 2 , wherein the determining of the feature set of the vehicle comprises:

securing data about the vehicle;

pre-processing and labeling the secured data;

training, based on the pre-processed and labeled data, the supervised learning-based artificial intelligence model; and

storing, in the database, a feature set generated by the training,

wherein the training of the supervised learning-based artificial intelligence model uses a semi-supervised learning methodology.

4 . The method of claim 3 , wherein the semi-supervised learning methodology comprises a MixMatch algorithm.

5 . The method of claim 1 , wherein the portion of the vehicle information of the vehicle comprises at least one of a transmission model, motor usage, an engine fuel, an engine displacement, or an engine type.

6 . The method of claim 1 , wherein the evaluation index is based on a confusion matrix-based performance evaluation index indicating, using an accuracy rating and a false positive (FP) rate, performance of the supervised learning-based artificial intelligence model.

7 . The method of claim 6 , wherein the determining of the feature set of the vehicle comprises:

selecting, among the candidate feature sets having accuracy ratings above a second threshold value and FP rates below a third threshold value, the candidate feature set based on the candidate feature set having a highest accuracy rating.

8 . The method of claim 7 , further comprising determining the accuracy rating of the confusion matrix-based performance evaluation index based on a formula, (TP+TN)/(TP+TN+FN+FP), and calculating an FP rate of the confusion matrix-based performance evaluation index based on a formula, FP/(TP+TN+FN+FP),

wherein TP represents true positive indicating that a positive prediction is correct, TN represents true negative indicating that a negative prediction is correct, FP represents false positive indicating that a positive prediction is wrong, and FN represents false negative indicating that a negative prediction is wrong, with respect to a transmission shift comfort level associated with the supervised learning-based artificial intelligence model.

9 . The method of claim 1 , further comprising:

causing a transmission diagnostic device to perform a diagnosis of the transmission of the vehicle using the output feature set of the vehicle.

10 . A method for diagnosing a transmission based on artificial intelligence, the method comprising:

measuring first transmission shift data of a vehicle;

pre-processing the measured first transmission shift data;

determining, using a supervised learning-based artificial intelligence model and based on a first feature set associated with the vehicle, a transmission shift comfort level;

receiving information on dissatisfaction of a customer regarding the transmission shift comfort level;

receiving, based on a determination that the information on the dissatisfaction indicates that the customer has a complaint associated with the determined transmission shift comfort level, customer complaint data comprising the pre-processed first transmission shift data and the determined transmission shift comfort level;

labeling misdiagnosed data confirmed by a user, based on an accumulated count of the received customer complaint data being above a first threshold value;

generating second transmission shift data by adding the misdiagnosed data to the first feature set;

training, based on the second transmission shift data, the supervised learning-based artificial intelligence model and generating a second feature set associated with the vehicle; and

storing the second feature set based on an evaluation index of the trained supervised learning-based artificial intelligence model being above a second threshold value.

11 . The method of claim 10 , wherein the generating of the second transmission shift data comprises:

determining a label ratio of the second transmission shift data;

training the supervised learning-based artificial intelligence model based on the label ratio satisfying a third threshold value; and

terminating the training of the supervised learning-based artificial intelligence model based on the label ratio not satisfying the third threshold value.

12 . The method of claim 10 , wherein the evaluation index is based on a confusion matrix-based performance evaluation index indicating, using an accuracy rating and a false positive (FP) rate, performance of the supervised learning-based artificial intelligence model.

13 . The method of claim 12 , wherein the storing of the second feature set comprises:

setting the second feature set of the trained supervised learning-based artificial intelligence model as a representative feature set of the vehicle based on the accuracy rating being above a fourth threshold value and the FP rate being below a fifth threshold value.

14 . An apparatus for diagnosing a transmission based on artificial intelligence, the apparatus comprising:

one or more processors; and

memory storing instructions that, when executed by the one or more processors, cause the apparatus to:

collect and pre-process vehicle information and transmission shift data associated with a vehicle;

label the transmission shift data;

among feature sets stored in a database, determine, as a plurality of candidate feature sets, a plurality of feature sets with respect to a plurality of other vehicles each associated with vehicle information that is same as at least a portion of the vehicle information of the vehicle;

generate a candidate feature set list comprising the plurality of candidate feature sets;

generate, using a supervised learning-based artificial intelligence model and for each of the plurality of candidate feature sets, an evaluation index of a transmission shift comfort level associated with the transmission shift data; and

output a feature set of the vehicle for diagnosing a transmission equipped in the vehicle by selecting, among candidate feature sets having evaluation indexes above a threshold value, a candidate feature set having a highest evaluation index.

15 . The apparatus of claim 14 , wherein the instructions, when executed by the one or more processors, further cause the apparatus to, based on none of a second plurality of candidate feature sets having an evaluation index above the threshold value, generate a new feature set based on additional transmission shift data associated with the vehicle.

16 . The apparatus of claim 15 , wherein the instructions, when executed by the one or more processors, cause the apparatus to generate the new feature set by using a semi-supervised learning methodology comprising a MixMatch algorithm.

17 . The apparatus of claim 14 , wherein the evaluation index is based on a confusion matrix-based performance evaluation index indicating, using an accuracy rating and a false positive (FP) rate, performance of the supervised learning-based artificial intelligence model; and

wherein the instructions, when executed by the one or more processors, cause the apparatus to determine the feature set of the vehicle by selecting, among the candidate feature sets having accuracy ratings above a second threshold value and FP rates below a third threshold value, the candidate feature set based on the candidate feature set having a highest accuracy rating.

18 . The apparatus of claim 14 , wherein the instructions, when executed by the one or more processors, further cause the apparatus to, based on customer complaint data associated with the transmission shift comfort level satisfying a first threshold value:

generate second transmission shift data by:

labeling misdiagnosed data confirmed by a user; and

adding the misdiagnosed data to existing data to generate an updated feature set; and

based on an evaluation index associated with the updated feature set satisfying a second threshold value, determine the updated feature set as a final feature set of the vehicle and storing the updated feature set in the database.

19 . The apparatus of claim 18 , wherein the instructions, when executed by the one or more processors, further cause the apparatus to:

use a confusion matrix-based performance evaluation index configured to evaluate performance of the supervised learning-based artificial intelligence model according to an accuracy rating and a false positive (FP) rate; and

determine the updated feature set as the final feature set of the vehicle, based on the accuracy rating of the confusion matrix-based performance evaluation index being above a third threshold value and the FP rate of the confusion matrix-based performance evaluation index being below a fourth threshold value.

20 . The apparatus of claim 14 , wherein the instructions, when executed by the one or more processors, further cause the apparatus to cause a transmission diagnostic device to perform a diagnosis of the transmission of the vehicle using the output feature set of the vehicle.