Distributed platform for network analysis

1. A method for a computer to generate and present data analysis to a user, comprising:

receiving a set of input files describing subject data;

generating a data model from the input files;

storing the subject data in the data model;

receiving an ontology comprising relationships between entities in the dataset and additional entities in an additional dataset of a previous data analysis model;

generating a data analysis model that at least includes a dataset from the data model, the ontology, and the additional dataset for analysis by a plurality of data analysis sub-algorithms of a data analysis algorithm, wherein the generating the data analysis model comprises at least one of calculating a probability that a topic corresponds to an input file or calculating a probability that an expression corresponds to the input file; and

performing an offline data analysis on the data analysis model that includes the dataset, the ontology, and the additional dataset, wherein the offline data analysis comprises:

selecting a sequence of data analysis sub-algorithms based on the data analysis model, each data analysis sub-algorithm to receive a corresponding input and performs statistical analysis on the corresponding input to produce a corresponding output,

specifying the dataset as an initial input to an initial data analysis sub-algorithm of the sequence,

pipelining an output produced by each data analysis sub-algorithm of the sequence as an input into each subsequent data analysis sub-algorithm, and

terminating the data analysis as a last data analysis sub-algorithm of the sequence produces a final output; and

presenting the final output on a data display.

2. The method of claim 1 , wherein the input files describe at least one of entities, attributes of the entities, or relationships between the entities, and wherein generating the data model from the input files comprises extracting at least one of the entities, the attributes of the entities, or the relationships between the entities from the input files.

3. The method of claim 1 , wherein the data analysis is network analysis, the data analysis model is a topic model, and the files are documents.

4. The method of claim 1 , wherein the pipelining comprises:

producing a specific output using a particular data analysis sub-algorithm,

determining whether the specific output of the particular data analysis sub-algorithm has a format that is compatible with a subsequent data analysis sub-algorithm in the sequence of data analysis sub-algorithms; and

in response to determining that the specified output of the particular data analysis sub-algorithm is incompatible with the subsequent data analysis sub-algorithm:

identifying a converter to convert the format of the specific output of the particular data analysis sub-algorithm to a converted format that is compatible with the subsequent data analysis sub-algorithm,

converting the specific output into a converted output having a converted format that is compatible with the subsequent data analysis sub-algorithm of the sequence of data analysis algorithms using the converter, and

inputting the converted output into the subsequent data analysis sub-algorithm.

5. The method of claim 1 , wherein the specifying the dataset as the initial input to the initial data analysis sub-algorithms includes specifying the initial input via a corresponding graphical user interface of the initial data analysis sub-algorithm.

6. The method of claim 1 , wherein specifying the dataset as the dataset as the initial input to a data analysis sub-algorithm comprises:

presenting at least some help text of a tutorial description of a data analysis sub-algorithm comprising one or more levels of detail to the user via an interactive user interface,

presenting parameter specifying user interface controls to specify the input and to specify a subsequent data analysis sub-algorithm to receive an output of the data analysis sub-algorithm, and

presenting help navigation user interface controls to enable the user to navigate through the one or more levels of detail of the tutorial description of the data analysis sub-algorithm.

7. The method of claim 1 , wherein at least some of the offline data analysis is performed on a cloud hosted resource.

8. The method of claim 1 , wherein the storing includes storing the subject data in the input files into one of a relational database, an object database, an object relational database, and a columnar database.

9. A computer-implemented system of one or more computing devices, the computer-implemented system comprising:

one or more processors; and

memory to store computer-executable instructions that, when executed by the one or more processors, perform acts comprises:

receiving a set of input files describing subject data;

generating a data model from the input files;

storing the subject data in the data model;

receiving an ontology comprising relationships between entities in the dataset and additional entities in an additional dataset of a previous data analysis model;

generating a data analysis model that at least includes a dataset from the data model, the ontology, and the additional dataset for analysis by a data analysis sub-algorithm of a data analysis algorithm, wherein the generating the data analysis model comprises at least one of calculating a probability that a topic corresponds to an input file or calculating a probability that an expression corresponds to the input file; and

performing an offline data analysis on the data analysis model that includes the dataset, the ontology, and the additional dataset, wherein the offline data analysis comprises:

selecting a sequence of data analysis sub-algorithms based on the data analysis model, the sequence of data analysis sub-algorithms to perform a statistical analysis on the data analysis model that at least provide a frequency of each data object in the dataset or correlation probabilities between data objects in the dataset, each data analysis sub-algorithm to receive a corresponding input and produce a corresponding output,

specifying the dataset as an initial input to an initial data analysis sub-algorithm of the sequence,

pipelining an output produced by each data analysis sub-algorithm of the sequence as an input into each subsequent data analysis sub-algorithm,

terminating the data analysis as a last data analysis sub-algorithm of the sequence produces a final output that includes a plurality of nodes associated with data objects connected by edges that represent relationships between the nodes; and

presenting the final output on a data display.

10. The computer-implemented system of claim 9 , wherein the input files describe at least one of entities, attributes of the entities, or relationships between the entities, and wherein generating the data model from the input files comprises extracting at least one of the entities, the attributes of the entities, or the relationships between the entities from the input files.

11. The computer-implemented system of claim 9 , wherein the data analysis is network analysis, the analysis model is a topic model, and the files are documents.

12. The computer-implemented system of claim 9 , wherein pipelining the at least one output of at least one specified data analysis sub-algorithm comprises:

producing a specific output using a particular data analysis sub-algorithm,

determining whether the specific output of the particular data analysis sub-algorithm has a format that is compatible with a subsequent data analysis sub-algorithm; and

in response to determining that the specified output of the particular data analysis sub-algorithm is incompatible with the subsequent data analysis sub-algorithm:

identifying a converter to convert the format of the specific output to a converted format that is compatible with the subsequent data analysis sub-algorithm,

converting the specific output into a converted output having a converted format that is compatible with the subsequent data analysis sub-algorithm using the converter, and

inputting the converted output into the subsequent data analysis sub-algorithm.

13. The computer-implemented system of claim 9 , wherein the specifying the dataset as the initial input to the initial data analysis sub-algorithms includes specifying the input via a corresponding graphical user interface of the initial data analysis sub-algorithm.

14. The computer-implemented system of claim 9 , wherein specifying the initial input to the initial data analysis sub-algorithm comprises:

presenting at least some help text of a tutorial description of the initial data analysis sub-algorithm comprising one or more levels of detail to the user via an interactive user interface,

presenting parameter specifying user interface controls to specify the initial input and to specify a subsequent data analysis sub-algorithm to receive an output of the initial data analysis sub-algorithm, and

presenting help navigation user interface controls to enable the user to navigate through the one or more levels of detail of the tutorial description of the initial data analysis sub-algorithm.

15. The computer-implemented system of claim 9 , wherein at least some of the offline data analysis is performed on a cloud hosted resource.

16. The computer-implemented system of claim 9 , The method of claim 1 , wherein the storing includes storing the subject data in the input files into one of a relational database, an object database, an object relational database, and a columnar database.