Systems and methods for identifying signals of low-quality listings

1. A computer-implemented method for identifying low-quality signals of an electronic listing of a listing entity, comprising:

ingesting, by one or more processors, text including (i) listing information of the electronic listing and (ii) listing entity information of the listing entity;

normalizing, by the one or more processors, the text by replacing first text of the text with second text from a list of predetermined text;

parsing, by the one or more processors, the text to identify a plurality of phrases included in the text;

generating, by the one or more processors, a respective set of category scores for each of at least one of the plurality of phrases by applying a scoring algorithm, wherein the respective set of category scores includes a plurality of scores for each of a plurality of low-quality categories;

identifying, by the one or more processors, one or more low-quality signals which each include:

(i) a respective low-quality phrase that is a phrase, of the at least one of the plurality of phrases, that corresponds to at least one score, of the respective set of low-quality category scores, that satisfies a low-quality threshold, and

(ii) at least one of the plurality of low-quality categories that corresponds to the at least one score that satisfies the low-quality threshold; and

presenting, by the one or more processors, an indication of the one or more low-quality signals together with the electronic listing to a user via a graphical user interface.

2. The method of claim 1 , further comprising:

generating, by the one or more processors, an aggregated listing entity score of the listing entity based on the one or more low-quality signals; and

indicating, by the one or more processors, the aggregated listing entity score.

3. The method of claim 1 , further comprising:

receiving, at the one or more processors, feedback from a user following presenting the indication of the one or more low-quality signals, the feedback regarding the low-quality signals and including user input;

analyzing, by the one or more processors, the feedback from the user to determine an inferred value of the indication of the one or more low-quality signals; and

updating, by the one or more processors, the scoring algorithm based on the feedback from the user.

4. The method of claim 1 , further comprising:

receiving, by the one or more processors, a training data set comprising a plurality of electronic listings; and

generating, by the one or more processors, the scoring algorithm by training a machine learning model using the training data set.

5. The method of claim 1 , wherein at least one of: (i) the first text includes an acronym and the second text does not include the acronym, or (ii) the first text includes a contraction and the second text does not include the contraction.

6. The method of claim 1 , wherein the low-quality categories are related to one or more of: deceptive pricing, deceptive condition, or deceptive inventory.

7. The method of claim 1 , further comprising:

in response to identifying the one or more low-quality signals, at least one of:

(i) generating and sending, by the one or more processors, correspondence regarding the one or more low-quality signals to the listing entity, or

(ii) applying, by the one or more processors, a modification to the electronic listing.

8. A computer system for identifying low-quality signals of an electronic listing of a listing entity, comprising:

one or more processors;

a program memory coupled to the one or more processors and storing executable instructions that, when executed by the one or more processors, cause the computer system to:

ingest text including: (i) listing information of the electronic listing, and (ii) listing entity information of the listing entity;

normalize the text by replacing first text of the text with second text from a list of predetermined text;

parse the text to identify a plurality of phrases included in the text;

generate a respective set of category scores for each of at least one of the plurality of phrases by applying a scoring algorithm, wherein the respective set of category scores includes a plurality of scores for each of a plurality of low-quality categories;

identify one or more low-quality signals which each include:

(i) a respective low-quality phrase that is a phrase, of the at least one of the plurality of phrases, that corresponds to at least one score, of the respective set of low-quality category scores, that satisfies a low-quality threshold, and

(ii) at least one of the plurality of low-quality categories that corresponds to the at least one score that satisfies the low-quality threshold; and

present an indication of the one or more low-quality signals together with the electronic listing to a user via a graphical user interface.

9. The computer system of claim 8 , wherein the executable instructions further cause the computer system to:

generate an aggregated listing entity score of the listing entity based on the one or more low-quality signals; and

indicate the aggregated listing entity score.

10. The computer system of claim 8 , wherein the executable instructions further cause the computer system to:

receive feedback from a user following presenting the indication of the one or more low-quality signals, the feedback regarding the low-quality signals and including user input;

analyze the feedback from the user to determine inferred value of the indication of the one or more low-quality signals; and

update the scoring algorithm based on the feedback from the user.

11. The computer system of claim 8 , wherein the executable instructions further cause the computer system to:

receive a training data set comprising a plurality of electronic listings; and

generate the scoring algorithm by training a machine learning model using the training data set.

12. The computer system of claim 8 , wherein at least one of: (i) the first text includes an acronym and the second text does not include the acronym, or (ii) the first text includes a contraction and the second text does not include the contraction.

13. The computer system of claim 8 , wherein the low-quality categories are related to one or more of: deceptive pricing, deceptive condition, or deceptive inventory.

14. The computer system of claim 8 , wherein the executable instructions further cause the computer system to:

in response to identifying the one or more low-quality signals, at least one of:

(i) generating and sending, by the one or more processors, correspondence regarding the one or more low-quality signals to the listing entity, or

(ii) applying, by the one or more processors, a modification to the electronic listing.

15. The computer system of claim 8 , wherein the executable instructions further cause the computer system to:

receive a training data set comprising a plurality of electronic listings; and

generate the scoring algorithm by training a machine learning model using the training data set.

16. A tangible, non-transitory computer-readable medium storing executable instructions for identifying low-quality signals of an electronic listing of a listing entity that, when executed by one or more processors of a computer system, cause the computer system to:

ingest text including one or more of: (i) listing information of the electronic listing, or (ii) listing entity information of the listing entity;

normalize the text by replacing first text of the text with second text from a list of predetermined text;

parse the text to identify a plurality of phrases included in the text;

generate a respective set of category scores for each of at least one of the plurality of phrases by applying a scoring algorithm, wherein the respective set of category scores includes a plurality of scores for each of a plurality of low-quality categories;

identify one or more low-quality signals which each include:

(i) a respective low-quality phrase that is a phrase, of the at least one of the plurality of phrases, that corresponds to at least one score, of the respective set of low-quality category scores, that satisfies a low-quality threshold, and

(ii) at least one of the plurality of low-quality categories that corresponds to the at least one score that satisfies the low-quality threshold; and

present an indication of the one or more low-quality signals together with the electronic listing to a user via a graphical user interface.

17. The tangible, non-transitory computer-readable medium of claim 16 , wherein:

the executable instructions further cause the computer system to:

generate an aggregated listing entity score of the listing entity based on the one or more low-quality signals; and

indicate the aggregated listing entity score.

18. The tangible, non-transitory computer-readable medium of claim 16 , wherein:

the executable instructions further cause the computer system to:

receive feedback from a user following presenting the indication of the one or more low-quality signals, the feedback regarding the low-quality signals and including user input;

analyze the feedback from the user to determine inferred value of the indication of the one or more low-quality signals; and

update the scoring algorithm based on the feedback from the user.

19. The tangible, non-transitory computer-readable medium of claim 16 , wherein:

the executable instructions further cause the computer system to:

receive a training data set comprising a plurality of electronic listings; and

generate the scoring algorithm by training a machine learning model using the training data set.

20. The tangible, non-transitory computer-readable medium of claim 16 , wherein:

the executable instructions further cause the computer system to:

in response to identifying the one or more low-quality signals, at least one of:

(i) generating and sending, by the one or more processors, correspondence regarding the one or more low-quality signals to the listing entity, or

(ii) applying, by the one or more processors, a modification to the electronic listing.