Data analysis system using artificial intelligence

1. A data analysis system utilizing custom unsupervised machine learning processes over a communications network, the system comprising:

a database connected to the communications network;

a web server comprising a memory and a processor, the web server connected to database and to the communications network;

a web application deployed on the web server, the web application including a data collection interface between the web server and the database of data, wherein the web application is configured for:

a) receiving input from a user via a graphical user interface, the input comprising modifying, by a user, a plurality of threshold parameters of a clustering algorithm for clustering a dataset from the database;

b) executing, on the dataset, a machine learning logistic regression function configured to calculate a metric of influence for each feature of a plurality of features associated with the dataset;

c) executing, on the dataset, the clustering algorithm with the plurality of threshold parameters that were modified by the user, thereby producing a current set of results comprising a plurality of scores associated with each feature of the plurality of features associated with the dataset;

d) identifying data in the dataset with a metric of influence and a plurality of scores below a predefined threshold, wherein said data that was identified is set for removal from the dataset;

e) removing the data that was identified from the dataset, thereby creating a revised dataset;

f) after producing the current set of results, providing a graphical user interface for the user to adjust the plurality of threshold parameters,

wherein if the user adjusts the plurality of threshold parameters, thereby producing an adjusted plurality of threshold parameters, then re-executing steps a) through f) using the adjusted plurality of threshold parameters and the revised dataset,

and wherein if the user does not adjust the plurality of threshold parameters, then the revised dataset is deemed a final dataset and the current set of results is deemed a final set of results;

g) copying the final set of results into a deep learning software framework; and

h) executing a deep learning algorithm in the deep learning software framework on the final set of results, thereby establishing relationships between the final dataset, and providing generalizations of the final dataset.

2. The data analysis system of claim 1 ,

wherein the clustering algorithm comprises at least a machine learning logistic regression function that is further configured for identifying a set of functional data comprised within the dataset.

3. The data analysis system of claim 2 , further comprising:

wherein the plurality of threshold parameters comprises a plurality of numerical values, wherein each numerical value comprises a decimal number.

4. The data analysis system of claim 3 , further comprising:

wherein the graphical user interface for the user to adjust the plurality of threshold parameters comprises a supportive graphical user interface.

5. The data analysis system of claim 4 , further comprising:

wherein the web application is further configured for generating a downloadable report comprising the final set of results for review by the user.

6. A method for data analysis utilizing custom unsupervised machine learning processes over a communications network, the method comprising:

storing data in a database connected to the communications network;

providing a web server comprising a memory and a processor, the web server connected to database and to the communications network;

providing a web application deployed on the web server, the web application including a data collection interface between the web server and the database, wherein the web application is configured for:

a) receiving input from a user via a graphical user interface, the input comprising modifying, by a user, a plurality of threshold parameters of a clustering algorithm for clustering a dataset from the database;

b) executing, on the dataset, a machine learning logistic regression function configured to calculate a metric of influence for each feature of a plurality of features associated with the dataset;

c) executing, on the dataset, the clustering algorithm with the plurality of threshold parameters that were modified by the user, thereby producing a current set of results comprising a plurality of scores associated with each feature of the plurality of features associated with the dataset;

d) identifying data in the dataset with a metric of influence and a plurality of scores below a predefined threshold, wherein said data that was identified is set for removal from the dataset;

e) removing the data that was identified from the dataset, thereby creating a revised dataset;

f) after producing the current set of results, providing a graphical user interface for the user to adjust the plurality of threshold parameters,

wherein if the user adjusts the plurality of threshold parameters, thereby producing an adjusted plurality of threshold parameters, then re-executing steps a) through f) using the adjusted plurality of threshold parameters and the revised dataset,

and wherein in response to a user review of the current set of results, setting the revised dataset as a final dataset and setting the current set of results as a final set of results;

g) copying the final set of results into a deep learning software framework; and

h) executing a deep learning algorithm in the deep learning software framework on the final set of results, thereby establishing relationships between the final dataset, and providing generalizations of the final dataset.

7. The method of claim 6 , further comprising:

wherein the clustering algorithm comprises at least a machine learning logistic regression function that is further configured for identifying a set of functional data comprised within the dataset.

8. The data method of claim 7 , further comprising:

wherein the plurality of threshold parameters comprises a plurality of numerical values, wherein each numerical value comprises a decimal number.

9. The method of claim 8 , further comprising:

wherein the graphical user interface for the user to adjust the plurality of threshold parameters comprises a supportive graphical user interface.

10. The method of claim 9 , further comprising:

wherein the web application is further configured for generating a downloadable report comprising the final set of results for review by the user.