Incorporating data into search engines using deep learning mechanisms

1. A computer-implemented method comprising:

generating one or more data sources pertaining to technology-related products based at least in part on processing user reviews pertaining to one or more technology-related products, call data related to interactions with one or more users pertaining to one or more technology-related products, and one or more technology-related product catalogs;

extracting, using one or more machine learning algorithms, one or more features from the one or more data sources pertaining to technology-related products by dividing text from the one or more data sources pertaining to technology-related products into a plurality of word portions, at least some of the plurality of word portions being of equal length, and wherein using the one or more machine learning algorithms comprises generating an output by processing the plurality of word portions using at least one machine learning hashing algorithm comprising a predetermined number of bins, wherein using the at least one machine learning hashing algorithm comprises (i) converting, using at least a portion of the predetermined number of bins, at least a portion of characters present in the word portions into a number within a predetermined range of values, (ii) converting the word portions to hash values, and (iii) associating the hash values with the corresponding numbers within the predetermined range of values associated with the at least a portion of characters in the given word portions;

generating one or more word vectors by applying at least one deep learning technique to the one or more extracted features, wherein applying the at least one deep learning technique to the one or more extracted features comprises determining one or more hierarchical relationships among at least a portion of the one or more extracted features by processing the generated output of the at least one machine learning hashing algorithm using at least one neural network;

mapping, using at least one deep similarity network, at least a portion of the one or more generated word vectors to one or more portions of text derived from a search query, wherein the search query comprises an e-commerce search query; and

outputting, based at least in part on the mapping, one or more results in response to the search query;

wherein the method is performed by at least one processing device comprising a processor coupled to a memory.

2. The computer-implemented method of claim 1 , wherein the one or more machine learning algorithms comprise one or more natural language processing algorithms.

3. The computer-implemented method of claim 1 , wherein the one or more machine learning algorithms comprises a latent Dirichlet allocation (LDA).

4. The computer-implemented method of claim 1 , wherein the search query comprises a text-based search query.

5. The computer-implemented method of claim 1 , wherein the search query comprises a voice-based search query.

6. The computer-implemented method of claim 2 , wherein the one or more natural language processing algorithms comprise at least one neural network model.

7. The computer-implemented method of claim 6 , wherein the at least one neural network model comprises at least one word2vec neural network model.

8. The computer-implemented method of claim 1 , further comprising:

determining a measure of similarity between the one or more generated word vectors and the one or more portions of text derived from the search query.

9. The computer-implemented method of claim 8 , wherein the measure of similarity comprises a cosine similarity.

10. A non-transitory processor-readable storage medium having stored therein program code of one or more software programs, wherein the program code when executed by at least one processing device causes said at least one processing device:

to generate one or more data sources pertaining to technology-related products based at least in part on processing user reviews pertaining to one or more technology-related products, call data related to interactions with one or more users pertaining to one or more technology-related products, and one or more technology-related product catalogs;

to extract, using one or more machine learning algorithms, one or more features from the one or more data sources pertaining to technology-related products by dividing text from the one or more data sources pertaining to technology-related products into a plurality of word portions, at least some of the plurality of word portions being of equal length, and wherein using the one or more machine learning algorithms comprises generating an output by processing the plurality of word portions using at least one machine learning hashing algorithm comprising a predetermined number of bins, wherein using the at least one machine learning hashing algorithm comprises (i) converting, using at least a portion of the predetermined number of bins, at least a portion of characters present in the word portions into a number within a predetermined range of values, (ii) converting the word portions to hash values, and (iii) associating the hash values with the corresponding numbers within the predetermined range of values associated with the at least a portion of characters in the given word portions;

to generate one or more word vectors by applying at least one deep learning technique to the one or more extracted features, wherein applying the at least one deep learning technique to the one or more extracted features comprises determining one or more hierarchical relationships among at least a portion of the one or more extracted features by processing the generated output of the at least one machine learning hashing algorithm using at least one neural network;

to map, using at least one deep similarity network, at least a portion of the one or more generated word vectors to one or more portions of text derived from a search query, wherein the search query comprises an e-commerce search query; and

to output, based at least in part on the mapping, one or more results in response to the search query.

11. The non-transitory processor-readable storage medium of claim 10 , wherein the one or more machine learning algorithms comprise one or more natural language processing algorithms.

12. The non-transitory processor-readable storage medium of claim 11 , wherein the one or more natural language processing algorithms comprise at least one neural network model.

13. The non-transitory processor-readable storage medium of claim 10 , wherein the one or more machine learning algorithms comprises a latent Dirichlet allocation (LDA).

14. The non-transitory processor-readable storage medium of claim 10 , wherein the search query comprises one of a text-based search query and a voice-based search query.

15. An apparatus comprising:

at least one physical processing device comprising a physical processor coupled to a memory;

the at least one physical processing device being configured:

to generate one or more data sources pertaining to technology-related products based at least in part on processing user reviews pertaining to one or more technology-related products, call data related to interactions with one or more users pertaining to one or more technology-related products, and one or more technology-related product catalogs;

to extract, using one or more machine learning algorithms, one or more features from the one or more data sources pertaining to technology-related products by dividing text from the one or more data sources pertaining to technology-related products into a plurality of word portions, at least some of the plurality of word portions being of equal length, and wherein using the one or more machine learning algorithms comprises generating an output by processing the plurality of word portions using at least one machine learning hashing algorithm comprising a predetermined number of bins, wherein using the at least one machine learning hashing algorithm comprises (i) converting, using at least a portion of the predetermined number of bins, at least a portion of characters present in the word portions into a number within a predetermined range of values, (ii) converting the word portions to hash values, and (iii) associating the hash values with the corresponding numbers within the predetermined range of values associated with the at least a portion of characters in the given word portions;

to generate one or more word vectors by applying at least one deep learning technique to the one or more extracted features, wherein applying the at least one deep learning technique to the one or more extracted features comprises determining one or more hierarchical relationships among at least a portion of the one or more extracted features by processing the generated output of the at least one machine learning hashing algorithm using at least one neural network;

to map, using at least one deep similarity network, at least a portion of the one or more generated word vectors to one or more portions of text derived from a search query, wherein the search query comprises an e-commerce search query; and

to output, based at least in part on the mapping, one or more results in response to the search query.

16. The apparatus of claim 15 , wherein the one or more machine learning algorithms comprise one or more natural language processing algorithms.

17. The apparatus of claim 16 , wherein the one or more natural language processing algorithms comprise at least one neural network model.

18. The apparatus of claim 15 , wherein the one or more machine learning algorithms comprises a latent Dirichlet allocation (LDA).

19. The apparatus of claim 15 , wherein the search query comprises a text-based search query.

20. The apparatus of claim 15 , wherein the search query comprises a voice-based search query.