Method and system for semantic analysis of unstructured data

1. A computer implemented method for semantically analyzing unstructured data, wherein the unstructured data is customer feedback comprising text for reviewing at least one of a product or service, the method comprising:

extracting data related to customer feedback from one or more digital input sources;

monitoring the one or more digital input sources to detect a plurality of parameters corresponding to each of said one or more digital input sources, wherein the detected plurality of parameters include business alliance and proximity, frequency of data updates, historical errors, data size, and number of concurrent system or user accesses corresponding to the digital input sources;

assigning weights to the one or more digital input sources, using a processor, wherein assignment of weights is based on the detected plurality of parameters corresponding to each of the one or more digital input sources;

extracting relevant text from the feedback data;

detecting and annotating sentence types from the relevant text;

determining relevant adjectives and associating them with the sentence types;

calculating a rating for each sentence of the relevant text using at least the weight assigned to the corresponding digital input source; and

providing an output in a pre-determined format.

2. The method of claim 1 , wherein the step of extracting data related to customer feedback further comprises crawling through digital input sources to obtain text related to customer feedback, wherein the one or more digital input sources include at least one of indicated data sources and non-designated data sources.

3. The method of claim 1 , wherein the step of extracting relevant text further comprises:

stripping out irrelevant text from the customer data;

staging the stripped text in a database; and

cleansing the data to obtain the relevant text.

4. The method of claim 1 , wherein the step of detecting and annotating sentence types further comprises the steps of:

detecting the type of sentence, wherein the type of sentence comprises at least one of an affirmative sentence, an interrogative sentence and a negative sentence; and

adding tags to each type of sentence to identify parts of speech.

5. The method of claim 1 further comprising using grammatical rules, linguistic patterns and corpus of text models to determine relevant adjectives and associating them with the sentence types.

6. The method of claim 5 , wherein the step of calculating a rating for each sentence of the relevant text further comprises using a rating model for assigning ratings based on the relevant adjectives and the sentence types.

7. The method of claim 6 further comprising combining the results of the rating model with previously determined numeric ratings available from external websites to calculate a collective rating.

8. The method of claim 7 further comprising combining the use of the rating model with a configurable scale, wherein the configurable scale is configured by assigning weights to one or more credibility parameters of pre-rated reviews.

9. The method of claim 6 , further comprising normalizing the results of the rating model by calculating a Bayesian average for the results.

10. A computer program product comprising a non-transitory computer usable medium having a computer readable program code embodied thereon for semantically analyzing unstructured data, wherein the unstructured data is customer feedback comprising text for reviewing at least one of a product or service, the computer program product comprising instructions that, when executed by a processor, cause the processor to perform a method comprising:

extracting data related to customer feedback from one or more digital input sources;

monitoring the one or more digital input sources to detect a plurality of parameters corresponding to each of said digital input sources, wherein the detected plurality of parameters include business alliance and proximity, frequency of data updates, historical errors, data size, and number of concurrent system or user accesses corresponding to the one or more digital input sources;

assigning weights to the one or more digital input sources, wherein assignment of weights is based on the detected plurality of parameters corresponding to each of the one or more digital input sources;

extracting relevant text from the feedback data;

detecting and annotating sentence types from the relevant text;

determining relevant adjectives and associating them with the sentence types;

calculating a rating for each sentence of the relevant text using at least the weight assigned to the corresponding digital input source; and

providing an output in a pre-determined format.

11. The computer program product of claim 10 , wherein extracting data related to customer feedback comprises crawling through digital input sources to obtain text related to customer feedback, wherein the one or more digital input sources include at least one of indicated data sources and non-designated data sources.

12. The computer program product of claim 10 , wherein the extracting relevant text further comprises:

stripping out irrelevant text from the customer data;

staging the stripped text in a database; and

cleansing the data to obtain the relevant text.

13. The computer program product of claim 10 , wherein detecting and annotating sentence types further comprises:

detecting the type of sentence, wherein the type of sentence comprises at least one of an affirmative sentence, an interrogative sentence and a negative sentence; and

adding tags to each type of sentence to identify parts of speech.

14. The computer program product of claim 10 further comprising using grammatical rules, linguistic patterns and corpus of text models to determine relevant adjectives and associating them with the sentence types.

15. The computer program product of claim 14 , wherein calculating a rating for each sentence of the relevant text further comprises using a rating model for assigning ratings based on the relevant adjectives and the sentence types.

16. The computer program product of claim 15 further comprising combining the results of the rating model with previously determined numeric ratings available from the external websites to calculate a collective rating.

17. The computer program product of claim 16 further comprising combining the use of the rating model with a configurable scale, wherein the configurable scale is configured by assigning weights to one or more credibility parameters of pre-rated reviews.

18. The computer program product of claim 16 , further comprising normalizing the results of the rating model by calculating a Bayesian average for the results.