Systems and methods for formatting data using a recurrent neural network

1. A system for formatting data, the system comprising:

at least one memory storing instructions; and

one or more processors configured to execute the instructions to perform operations comprising:

receiving data comprising a plurality of data value sequences;

training a recurrent neural network model to output conditional probabilities of subsequent data values based on preceding data values in the received data value sequences;

generating conditional probabilities using the trained recurrent neural network model and the received data value sequences;

determining a data format of a subset of the received data value sequences, based on the generated conditional probabilities;

generating reformatted data based on the received data, wherein generating reformatted data comprises reformatting at least one of the received data value sequences according to the determined data format; and

training a synthetic data model to generate synthetic data using the reformatted data.

2. The system of claim 1 , wherein:

the operations further comprise generating embedded data based on the received data value sequences of data values; and

training the recurrent neural network model comprises using the embedded data as training data.

3. The system of claim 2 , wherein generating embedded data comprises implementing at least one of a one-hot encoding method or a glove method.

4. The system of claim 2 , wherein generating embedded data comprises implementing at least one of an autoencoder model, a transformer model, or an attention network model.

5. The system of claim 1 , wherein the operations further comprise displaying a probabilistic graph of the generated conditional probabilities.

6. The system of claim 1 , wherein the operations further comprise determining a frequency of the determined data format.

7. The system of claim 6 , wherein the operations further comprise displaying the frequency of the determined data format in a probabilistic graph of the generated conditional probabilities.

8. The system of claim 1 , wherein the operations further comprise at least one of storing or transmitting the reformatted sequence of data.

9. The system of claim 1 , wherein the operations further comprise updating a relational database based on the reformatted sequence of data.

10. The system of claim 1 , wherein the operations further comprise generating, based on the conditional probabilities, an expression for reformatting data.

11. The system of claim 10 , wherein the expression comprises a regex function.

12. The system of claim 10 , wherein the operations further comprise updating a library of functions to include the expression.

13. The system of claim 1 , wherein the operations further comprise:

generating a sequence of classifications of data values corresponding to at least one of the received data value sequences; and

the determined data format comprises the sequence of classifications of data values.

14. The system of claim 1 , wherein generating conditional probabilities comprises skipping a received data value sequence based on a timeout window.

15. The system of claim 1 , wherein:

training the recurrent neural network model further comprises training the recurrent neural network model to determine a data format of subsets of the received data value sequences based on the conditional probabilities; and

determining the data format comprises using the recurrent neural network model.

16. The system of claim 1 , wherein:

the recurrent neural network model is a first recurrent neural network model;

the operations further comprise training a second recurrent neural network model to determine a plurality of data formats based on the conditional probabilities, the plurality of data formats comprising the determined data format; and

determining the data format of the subset of the received data value sequences comprises using the second recurrent neural network model.

17. The system of claim 1 , wherein the received data value sequences are instances of a data type.

18. The system of claim 1 , wherein training the recurrent neural network comprises:

determining that a second data value sequence matches a first data value sequence previously processed by the recurrent neural network; and

skipping the second data value sequence.

19. A method for formatting data, the method comprising:

receiving data comprising a plurality of data value sequences;

training a recurrent neural network model to output conditional probabilities of subsequent data values, based on preceding data values in the received data value sequences;

determining a data format of the received data value sequences based on the conditional probabilities;

generating reformatted data based on the received data, wherein generating reformatted data comprises reformatting at least one of the received data value sequences according to the determined data format; and

providing the reformatted data to a synthetic data model, the synthetic data model being trained to generate synthetic data using training data comprising a data value sequence having the determined data format.

20. A system for formatting data, the system comprising:

at least one memory storing instructions; and

one or more processors configured to execute the instructions to perform operations comprising:

receiving data comprising a plurality of data value sequences;

generating embedded data based on the received data;

training a recurrent neural network model to output conditional probabilities of subsequent data values based on preceding data values in the received data value sequences, wherein training comprises using the embedded data as training data;

determining a plurality of data formats of the received data value sequences based on the conditional probabilities;

displaying the data formats at a user interface in a probabilistic graph;

receiving a selected data format from the user interface from among the displayed data formats;

generating reformatted data based on the received data, wherein generating reformatted data comprises reformatting at least one of the received data value sequences according to the selected data format; and

training a synthetic data model to generate synthetic data using the reformatted data.