Systems and methods for structuring data from unstructured electronic data files

Abstract

Computer implemented systems and methods are disclosed for structuring data from unstructured electronic data files. In accordance with some embodiments, an electronic data file including unstructured content associated with a legal process return is received and the unstructured content parsed. The unstructured content is parsed to identify one or more objects and properties based on a database ontology that are processed to generate an object model. A data report may be generated based on the identified objects and properties.

Description

BACKGROUND

Law enforcement agencies increasingly rely on social media data to perform criminal investigations. An agency typically serves a search warrant, national security letter, subpoena, or another type of legal process on a social media platform administrator which provides a legal process return to the agency in response to the legal process. Legal process returns may be provided as electronic data files in a number of formats including, for example, PDF files, text files, spreadsheets, and database files. They can include information such as, for example, contact information, friend lists, private messages, public posts, “tag” and “like” or “favourite” history, phone numbers, login history, and IP address information.

Problems arise when a legal process return is received as an electronic data file that includes unstructured data. The unstructured data, for example, may need to be manually processed by law enforcement agencies in order to aggregate the data and produce useful reports. Such manual processing may require significant amounts of time to accomplish (e.g., weeks or months) and can reduce the value of the acquired information, as the information may become stale or irrelevant during that time. Moreover, the size of unstructured electronic data files can make it difficult or impossible to view the files using native files viewers. For example, legal process returns that include unstructured data can include several hundreds of thousands of pages of data. These electronic data files may exceed sizes of 500 Mb, making it impossible for agencies to view and search the files on conventional data management systems.

BRIEF DESCRIPTION OF THE DRAWINGS

Reference will now be made to the accompanying drawings, which illustrate exemplary embodiments of the present disclosure and in which:

FIG. 1 is a block diagram of an exemplary system for structuring data from unstructured electronic data files, consistent with embodiments of the present disclosure.

FIG. 2 is a block diagram of an exemplary data structuring system for structuring data from unstructured electronic data files, consistent with embodiments of the present disclosure.

FIG. 3 illustrates an example object model, consistent with embodiments of the present disclosure.

FIG. 4 illustrates and example implementation of an interactive GUI, consistent with embodiments of the present disclosure.

FIGS. 5-7B illustrate embodiments of example data reports generated by the exemplary data structuring system of FIG. 2, consistent with embodiments of the present disclosure.

FIG. 8 is a flow diagram depicting an example method for structuring data from unstructured electronic data files, consistent with embodiments of the present disclosure.

DETAILED DESCRIPTION OF EXEMPLARY EMBODIMENTS

Reference will now be made in detail to exemplary embodiments, the examples of which are illustrated in the accompanying drawings. Whenever possible, the same reference numbers will be used throughout the drawings to refer to the same or like parts.

The disclosed embodiments describe improved methods and systems for structuring data from unstructured electronic data files. The improved data structuring systems and methods can receive electronic data files including unstructured social media content in excess of 500 Mb in size, parse the unstructured content, structure the parsed content by assigning object types and property types to the parsed content, and stored the structured content in a database. The disclosed data structuring systems and methods may aggregate the structured content to generate various types of data reports. The reports may include, for example, reconstructed conversations between a subject and their contacts, a list of normalized phone numbers associated with the subject, a geographic mapping of IP addresses associated with the subject, a list of IP addresses shared between the subject and other persons, a timeline of specific events (logins, subject movement, etc.), and other reports. The data structuring systems and methods may also present the aggregated structured content in an interactive graphical user interface that allows for free-form customization and exploration of the aggregated structured content.

Accordingly, the systems and methods described herein are capable of filtering large amounts of data in a quick, logical, and visually associative way. More specifically, the systems and methods can, among other things, provide the ability to display information about events and entities both temporally and geographically, and allow for the selection and grouping of different entities and events on the graphical representation. Furthermore, the disclosed systems and methods are capable of resolving multiple instances of object and property references across enterprise databases into a canonical format based on a database ontology.

FIG. 1 is a block diagram of an exemplary system environment 100 for structuring data from unstructured electronic data files, consistent with embodiments of the present disclosure. As shown in FIG. 1, system environment 100 includes a number of components. It will be appreciated from this disclosure, however, that the number and arrangement of these components is exemplary only and provided for purposes of illustration. Other arrangements and numbers of components may be utilized without departing from the teachings and embodiments of the present disclosure.

As shown in the example embodiment of FIG. 1, system environment 100 may include one or more social media platforms 110, 120. Social media platform 110, 120 may include platforms such as, for example, Facebook, Twitter, Instagram, SureSpot, Kik, PalTalk, or any other social media platform known in the art. Social media platform 110, 120 may be implemented by, for example, a server, a server system comprising a plurality of servers, a server farm comprising a load balancing system and a plurality of servers, a mainframe computer, or any combination of these components. In certain embodiments, social media platform 110, 120 may be a standalone computing system or apparatus, or it may be part of a subsystem, which may be part of a larger system. For example, social media platform 110, 120 may represent distributed servers that are remotely located and communicate over a communications medium (e.g., network 150) or over a dedicated network, for example, a LAN. In some embodiments, social media platform 110, 120 may be implemented with hardware devices and/or software applications running thereon. In some embodiments, social media platform 110, 120 may be configured to communicate to and/or through network 150 with other components such as data structuring system 130 and database 140, and vice-versa. Also, in some embodiments, social media platform 110, 120 may implement aspects of the present disclosure without the need for accessing another device, component, or network, such as network 150.

Network 150 may include any combination of communications networks. For example, network 150 may include the Internet and/or any type of wide area network, an intranet, a metropolitan area network, a local area network (LAN), a wireless network, a cellular communications network, etc. In some embodiments, client 110, 120 may be configured to transmit data and information through network 150 to an appropriate data importer, such as, for example, data importer 130. For example, client 110, 120 may be configured to transmit electronic data files including various types of content to data importer 130. In some aspects, client 110, 120 may also be configured to receive information from data importer 130 through network 150.

Data structuring system 130 may be configured to communicate and interact with social media platform 110, 120, and database 140. In certain embodiments, data structuring system 130 may be standalone system or apparatus, or it may be part of a subsystem, which may be part of a larger system. For example, data structuring system 130 may represent a distributed system that includes remotely located sub-system components that communicate over a communications medium (e.g., network 150) or over a dedicated network, for example, a LAN.

In some embodiments, data structuring system 130 may be configured to receive data and information through network 150 from various devices and systems, such as, for example, social media platform 110, 120. For example, data structuring system 130 may be configured to receive legal process returns in the form of electronic data files from social media platform 110, 120, and other devices and systems. The electronic data files may be received in various file formats and may include content that is provided by social media platform 110, 120 in response to a legal process such as warrant, national security letter, subpoena, etc., relating to a criminal investigation conducted by a law enforcement agency. The content may include social media content associated with a subject of the criminal investigation such as, for example, contact information, friend lists, private messages, phone numbers, login information, IP address information, photos, photo albums, profiles of persons associated with the subject, email addresses, public social media posts (e.g., wall posts, microblog posts such as Tweets, and status updates), location updates (e.g., check-ins and public posts regarding the subject's location), etc. Data structuring system 130 may be configured to structure and import the content included in the received electronic data files into one or more structured databases such as, for example, database 140.

Database 140 may include one or more logically and/or physically separate databases configured to store data. The data stored in database 140 may be received from data structuring system 130, from social media platform 110, 120 and/or may be provided as input using conventional methods (e.g., data entry, data transfer, data uploading, etc.). The data stored in the database 140 may take or represent various forms including, but not limited to, electronic data files, object mappings, property mappings, report templates, user profile information, and a variety of other electronic data or any combination thereof. In some embodiments, database 140 may include separate databases that store electronic data files, object and property mappings, and report templates, respectively. In still some other embodiments, the databases that store electronic data files, object and property mappings, and report templates can be combined into various combinations. In still some other embodiments, database 140 includes a single database that stores electronic data files, object and property mappings, and report templates.

In some embodiments, database 140 may be implemented using any suitable form of a computer-readable storage medium. In some embodiments, database 140 may be maintained in a network attached storage device, in a storage area network, or combinations thereof, etc. Furthermore, database 140 may be maintained and queried using numerous types of database software and programming languages, for example, SQL, MySQL, IBM DB2®, Microsoft Access®, PERL, C/C++, Java®, etc. Although FIG. 1 shows database 140 associated with data structuring system 130, database 140 may be a standalone database that is accessible via network 150, database 140 may be included in data structuring system 130, or database 140 may be associated with or provided as part of a system or environment that may be accessible to social media platform 110, 120 and/or other components.

FIG. 2 is a block diagram of an exemplary data structuring system 130 for implementing embodiments and aspects of the present disclosure. For example, data structuring system 130 may be used for structuring data from unstructured electronic data files. The arrangement and number of components included in data structuring system 130 is provided for purposes of illustration. Additional arrangements, number of components, and other modifications may be made, consistent with the present disclosure.

As shown in FIG. 2, data structuring system 130 may include one or more communications interfaces 210. Communications interface 210 may allow data and/or information to be transferred between data structuring system 130 and network 150, social media platform 110, 120, database 140, and/or other components. For example, communications interface 210 may be configured to receive legal process returns in the form of electronic data files that include unstructured content. Some non-limiting examples of electronic data files include word processing files (.pdf, .doc, .docx, .txt, .log, .rtf, etc.), spreadsheets (.xls, .xlsx, .ods, etc.), comma separated values (CSV) files, presentations, archived and compressed files (e.g., ZIP files, 7z files, cab files, RAR files, etc.), database files. PDF files, PUB files, image files, XML files, specialized tax and financial files (e.g., Open Financial Exchange and Interactive Financial Exchange files), tabulated data files and webpage files (e.g., HTML files). The received electronic data files may include various types of unstructured content. For example, the received electronic data files may include social media data associated with a subject of a criminal investigation as described above in reference to FIG. 1.

Examples of communications interface 210 may include a modem, a wired or wireless communications interface (e.g., an Ethernet, Wi-Fi, Bluetooth, Near Field Communication, WiMAX, WAN, LAN, etc.), a communications port (e.g., USB, IEEE 1394, DisplayPort, DVI, HDMI, VGA, Serial port, etc.), a PCMCIA slot and card, etc. Communications interface 210 may receive data and information in the form of signals, which may be electronic, electromagnetic, optical, or other signals capable of being received by communications interface 210. These signals may be provided to communications interface 210 via a communications path (not shown), which may be implemented using wireless, wire, cable, fiber optics, radio frequency (“RF”) link, and/or other communications channels.

Data structuring system 130 may also include one or more file databases 220. File database 220 may be configured to store electronic data files received by data structuring system 130 at communications interface 210.

Data structuring system 130 may also include one or more structuring components 230 that may parse the unstructured social media content included the electronic data files stored in file database 220 and structure the parsed data according to a database ontology 240. Exemplary embodiments for defining an ontology (such as database ontology 240) are described in U.S. Pat. No. 7,962,495 (the '495 Patent), issued Jun. 14, 2011, the entire contents of which are expressly incorporated herein by reference. Among other things, the '495 patent describes embodiments that define a dynamic ontology for use in creating data in a database. For creating a database ontology 240, for example, one or more object types may be created where each object type can include one or more properties. The attributes of object types or property types of the database ontology 240 can be edited or modified at any time.

In some embodiments, object types may be further divided into a number of sub-categories. For example, object types may be divided into entity types, event types and document types. Entity types may define a person, place, thing, or idea. Examples, of entity types include social media platform profile (e.g., Facebook™, or Twitter™ user profile), IP address, email address, photo album, friend's list, and location. Event types may define a type of social media platform event associated with the subject of a criminal investigation. Event types may include, for example, the subject logging into their social media platform profile, posting a photo to the subject's social media platform profile, sending friend requests, and accepting friend requests. Document types may define a type of social media platform document created by the subject or the subject's contacts. Examples of document types include private messages, status updates, microblog posts (e.g., Facebook™ wall posts Twitter™ Tweets), comments on other users' microblog posts, pictures, and videos.

In some embodiments, each property type is declared to be representative of one or more object types. A property type is representative of an object type when the property type is intuitively associated with the object type. For example, a property type of “Text/Description” may be representative of an object type “Private Message” but not representative of an object type “Photo Album.” In some embodiments, each property type has one or more components and a base type. In some embodiments, a property type may comprise a string, a date, a number, or a composite type consisting of two or more string, date, or number elements. Thus, property types are extensible and can represent complex data structures. Further, a parser definition can reference a component of a complex property type as a unit or token.

An example of a property having multiple components is a Name property having a Last Name component and a First Name component. An example of raw input data is “Smith, Jane.” An example parser definition specifies an association of imported input data to object property components as follows: {LAST_NAME}, {FIRST_NAME}→Name:Last, Name:First. In some embodiments, the association {LAST_NAME}, {FIRST_NAME} is defined in a parser definition using regular expression symbology. The association {LAST_NAME}, {FIRST_NAME} indicates that a last name string followed by a first name string comprises valid input data for a property of type Name. In contrast, input data of “Smith Jane” would not be valid for the specified parser definition, but a user could create a second parser definition that does match input data of “Smith Jane.” The definition Name:Last, Name:First specifies that matching input data values map to components named “Last” and “First” of the Name property. As a result, parsing the unstructured data in an electronic data file using the parser definition results in assigning the value “Smith” to the Name:Last component of the Name property, and the value “Jane” to the Name:First component of the Name property.

In some embodiments, object types and property types may be specific to each social media platform. For example, database ontology 240 may include sets of object types and property types that are specific to Facebook™, Twitter™, Instagram™, etc. In order to determine which set of object/property types to use for an electronic data file, structuring component 230 may scan a header included in the electronic data file to detect a social media platform identifier. For example, the header may include the name Facebook™ and the warrant or subpoena number. Structuring component 230 may detect the name Facebook™ in the file and select the set of Facebook™ object/property types in response.

In some embodiments, parser 232 may parse the unstructured content included in electronic data files stored in files database 220 to identify one or more objects based on the set of object/property types selected by structuring component 230. In order to parse the unstructured content, parser 232 may scan the unstructured content using natural language processing techniques to identify one or more words or strings of words. In some embodiments, where the electronic data files includes text that is unrecognizable by parser 232 (e.g., where the file includes PDF images of text), structuring component 230 may extract the text using techniques such as, for example, optical character recognition, optical word recognition, intelligent character recognition, and intelligent word recognition. Parser 232 may compare the identified words or strings of words to the selected set of object types defined in database ontology 240 to identify object types included in the electronic data file. Once an object type has been identified, parser 232 may identify objects included in the electronic data file of that object type. As an example, parser 232 may identify the string “Registered Email Address” and compare the string to object types defined in database ontology 240. If the string matches a known object type, parser 232 may identify the next string of text as the subject's email address (e.g., johndoe@email.com). A mapper 234 may assign object types and property types to the identified objects. The objects, assigned object types, and assigned property types make up a structured object model of the electronic data file. Each object model may correspond to a legal process return received in response to a legal process for social media platform content associated with a subject. The subject may be, for example, a subject of a criminal investigation conducted by a law enforcement agency. Object models may be stored in an object model database 250 and are described in more detail below in reference to FIG. 3.

In some embodiments, an object explorer 260 may generate an interactive graphical user interface (GUI) that allows for the customization and exploration of the structured objects and properties. For example, the interactive GUI may include various content filters that aggregate the structured objects and properties based on various filter properties. The content filters may, for example, filter objects based on entity type (e.g., IP address, email address, friend's list, etc.), event type (e.g., login events, phot post events, etc.), and document types (e.g., private message, social media profile status update, wall posts, etc.). The content filters may also filter properties based on, for example, property types (e.g., warrant number, online identifier, date range, location, etc.).

Once the structured objects and/or properties have been filtered based on one or more content filters, the interactive GUI may allow for customized data visualizations of the filtered data to be displayed. For example, a timeline of login events may be presented in the interactive GUI when the structured objects are filtered by a login event type. The timeline may display when the login events occurred. When unstructured content associated with multiple subjects have been structured and aggregated, the timeline presentation on the interactive GUI can display how many login events occurred at a given time and which subject logged in at a particular time so that conclusions about real-world interactions between the subjects can be deduced or inferred. In some embodiments, the customized data visualizations of the filtered data can be further customized, or a subset of the visualized data can be selected so that another customized data visualization can be displayed. For example, based on the login timeline example above, a subset of the visualized login data can be selected, geocoded (using a MaxMind database, for example), and used to generate a customized data visualization of a map showing the geographic locations associated with each selected login event. Accordingly, the interactive GUI allows for free-form interaction and customization of the structured objects and properties to generate useful visualizations of the structured objects and properties so that various conclusions and extrapolations can be performed.

As another example of the above interactive GUI, structured photograph objects may be filtered by a MD5 hash property type so that photograph objects stored in object model database 250 with same or similar MD5 hashes can be aggregated and their properties analysed. For example, a photograph with an MD5 hash may have been posted on a social media profile of a subject. The interactive GUI can filter structured photograph objects based on the MD5 hash of the posted photograph to identify other social media profiles associated with subjects that have also posted the same photograph, therefore allowing conclusions and inferences of interactions between subjects who have posted the same photograph to be drawn.

Object explorer may also generate various types of data reports based on the object models stored in object model database 250. The data reports may include data models of objects and properties defined in an object model such as, for example, timelines and geographic mappings of events, histograms of objects and properties, reconstruction of social media conversations (e.g., private message conversations between two or more users), mappings of shared IP addresses between two or more users, picture matching, friends list graphs, and other types of data models.

In order to generate a data report, object explorer may provide instructions to a GUI generator 290 to generate a GUI of object explorer 260. In response to the received instructions, GUI generator 290 may generate an interactive GUI for display on a display 295. Data structuring system 130 may also include one or more input/output (I/O) devices 270 (e.g., physical keyboards, virtual touch-screen keyboards, mice, joysticks, styluses, etc.) that are configured to receive user instructions in the form of user input. The received instructions may include instructions to generate data reports based on objected models stored in object model database 250. Object explorer 260 may receive the user input from I/O 270, generate the request data report based on a report template associated with the requested data report, and may provide instructions to GUI generator 290 for generating a display of the generated data report on display 295.

In some embodiments, object explorer 260 may include a template selector 262 that selects a report template among the report templates stored in a report template database 280. The template selection may be selected based on user input received from I/O 270. For example, the user input received at object explorer 260 may identify a data report type requested by the user, and template selector 262 may retrieve the report template corresponding to the requested data report type. As an example, if the user requests a data report of all the telephone numbers included in an object model, template selector 262 may select a telephone number histogram report template from report template database 280. As another example, if the user requests a data report including a geographic mapping of a subject's social media platform login activity between 10:30 p.m., Jul. 15, 2013 and 3:15 a.m., Jul. 16, 2013, template selector 262 may select the appropriate template from report template database 280.

Once template selector 262 has selected the appropriate report template for the requested data report, a template applicator 264 may obtain objects and properties included in the object model that are required by the report template. Template applicator 264 may generate the requested report using the obtained objects and properties based on the selected report template. Template applicator 264 may provide instructions for GUI generator 290 to display the generated data report on display 295.

Structuring component 230, object explorer 260, and GUI generator 290 may be implemented as hardware modules configured to execute the functions described herein. Alternatively, one or more processors suitable for the execution of instructions may be configured to execute the functions of structuring component 230, object explorer 260, and GUI generator 290. For example, suitable processors include both general and special purpose microprocessors, programmable logic devices, field programmable gate arrays, specialized circuits, and any one or more processors of any kind of digital computer that may be communicatively coupled to a physical memory (not shown) storing structuring component 230, object explorer 260, and GUI generator 290 in the form of instructions executable by the processor. Suitable memories may include, for example, NOR or NAND flash memory devices, Read Only Memory (ROM) devices, Random Access Memory (RAM) devices, storage mediums such as, for example, hard drives, solid state drives, tape drives, RAID arrays, etc. As another example, the functions of structuring component 230, object explorer 260, and GUI generator 290 may be included in the processor itself such that the processor is configured to implement these functions.

File database 220, database ontology 240, object model database 250, and report template database 280 may be implemented by database 140 of FIG. 1. In some embodiments, one or more of databases 220, 240, 250, and 280 may be included in the same database. In some embodiments, one or more of databases 220, 240, 250, and 280 may be included in separate databases.

Display 295 may be implemented using devices or technology, such as a cathode ray tube (CRT) display, a liquid crystal display (LCD), a plasma display, a light emitting diode (LED) display, a touch screen type display such as capacitive or resistive touchscreens, and/or any other type of display known in the art.

FIG. 3 is illustrative of an exemplary object model 300 and a corresponding ontology (e.g., database ontology 240 in FIG. 2). The elements of exemplary object model 300 can be stored in an object model database (e.g., object model database 250 of FIG. 2).

Object model 300 can include, among other things, entities 310A-C, event 320A, and documents 330A-B. Each entity 310, event 320, and document 330 can further contain properties including, without limitation, representative properties, base properties, or complex properties (e.g., transcript property 350A) made up of multiple sub properties or components. Complex properties can be used to provide detailed information about entities, events, and documents.

As illustrated in FIG. 3, entity 310A may correspond to a social media platform profile associated with the subject of a criminal investigation and entities 310B and 310C may correspond to social media platform profiles associated with persons with whom the subject of a criminal investigation has interacted. For example, the subject may have interacted with the associated persons via private message documents 330A and 330B.

Private message documents 330A and 330B may include various properties such as, for example, a transcript property, an IP address property, “TO” and “FROM” properties, and a “date/time” property. The transcript property, such as transcript property 350A, may contain the text of private message documents (e.g., private message document 330A) as well as additional properties. The additional properties may include, for example, the name of the transcript, the character count, read receipt information, telephone numbers included in the message, and/or any attachments in the message. For example, transcript property 350A may include telephone number property 350E, which may be assigned as a property of private message document 330A. In some embodiments, the transcript property could be in an audio format or some other format instead of written. It is appreciated that many different formats can be commonly used and would be known to one of ordinary skill in the art that could replace a written or audio property.

Additionally, events, documents, and entities can contain notes and media. Notes can provide a container for textual information related to the event, document, or entity. Media can represent binary data associated with the events, documents, or entities. Media data can take the form of, for example, text documents, images, videos, or specialized formats.

Moreover, both objects and properties can contain geospatial and temporal metadata. Geospatial metadata can provide a physical location associated with an object or property. For example, private message document 330A can have an IP address property 350B which can be used to obtain the geographic location of the subject associated with social media profile entity 330A that sent the private message. As another example, login event 320A can have an IP address property 350C associated with the person associated with social media profile entity 310A logging into a social media platform. It is appreciated that the geospatial data can also be in any form that represents a location and is understood by the users of object model 300. Temporal metadata can represent either a specific point in time or a duration having a start time and an end time. For example, private message document 330A can contain a “TIME” property 350D indicating a specific date and time when the message was sent. In some embodiments duration can be indicated by including a start property and end property allowing calculation of the duration. The temporal data can be in any form (e.g., epoch time, UTC time, or local time) that represents the time of the event or the duration of the event. Moreover, in some embodiments, geospatial and temporal metadata can be correlated. For example, the geospatial and temporal metadata can correspond to one or more locations and times when a person visited those one or more locations.

Entities 310, events 320, and documents 330 can serve as links indicating relationships between the various objects. For example, private message document 330A can contain “FROM” and “TO” properties. The “FROM” property links social media profile 310A to private message document 330A and the “TO” property links social media profile 310B to private message document 330A. Thus private message document 330A, while still containing its own relevant properties (e.g., temporal properties, geospatial properties, and transcript property 350A), can act as a complex link between social media profiles 310A and 310B.

FIG. 4 illustrates and example implementation of an interactive GUI 400 for free-form exploration of structured objects and properties. In some embodiments, example interactive GUI 400 may be generated by a data structuring system (e.g., data structuring system 130 including an object explorer 260, both of FIG. 2). GUI 400 may include a set of content filters such as, for example, object types 410 and property types 420. The object types 410 filter may further be divided into sub-filters such as, for example, entity types 412, event types 414, and document types 416. Content filters 410-416 and 420 are exemplary only and other filters may also be included in GUI 400. Content filters 410-416 and 420 allow for the aggregation of structured objects and properties so that customized data visualizations may be generated.

In some embodiments, GUI 400 may allow for customized data visualizations of data filtered by content filters 410-416 and 420 to be displayed. GUI 400 may include various visualization types 430 that can be used to generate displays of the filtered data. In the example illustrated in FIG. 4, a timeline visualization type, a pie chart visualization type, a histogram visualization type, and a bar chart visualization type are included in GUI 400. Other visualization types 430 and combinations of visualization types 430 may be included in GUI 400. In some embodiments, the visualization types 430 presented on GUI 400 may depend on the type of content filter selected. For example, if a login event type 416 filter is selected, GUI 400 may display a timeline visualization type (that displays the login events on a timeline), a histogram visualization type (that displays the number of login events associated with various IP addresses), and a pie chart visualization type.

A customized data visualization may be generated using various techniques. For example, input may be received (from I/O 270 of FIG. 2, for example) in the form of a selection of an object type 410 or a property type 420 and a visualization type 430. The input may be received in various forms. For example, the input may be a user selecting an object type 410 or a property type 420 and dragging it on top of a visualization type 430. As another example, the input may be a user highlighting an object type 410 or a property type 420 (by clicking on it, for example) and highlighting a visualization type 430.

In some embodiments, the customized data visualizations displayed on GUI 400 can be further customized, or a subset of the visualized data can be selected so that another customized data visualization can be displayed.

FIGS. 5-7B illustrate example implementations of data reports. In some embodiments, the example data reports may be generated by a data structuring system (e.g., data structuring system 130 including an object explorer 260, both of FIG. 2). FIG. 5 in particular illustrates an example implementation of a telephone number histogram data report 500. As shown in FIG. 5, data report 500 may include a list of telephone numbers 510A-D. Telephone numbers 510A-D may have been included in one or more private messages (e.g., private message document 330A of FIG. 3) between a subject of a criminal investigation and another person (e.g., John Doe, entity 310A, and Jane Smith, entity 310B, both of FIG. 3). A parser (e.g., parser 232 of FIG. 2) may have parsed the private messages to identify and normalize telephone numbers 510A-D to a telephone number format required by a database ontology (e.g., database ontology 240 of FIG. 2). As shown in data report 500, the data structuring system may represent the number of times a telephone number 510A-D has shown up in a private message between the subject and another person as a histogram. The histogram may include data bars 520A-D that graphically represent the number of times each telephone number 510A-D has shown up in a private message. The histogram may also include a numeric representation of the number of times each telephone number 510A-D has shown up in a private message proximate to data bars 520A-D. In some embodiments, and as shown in FIG. 5, telephone numbers 510A-D (and data bars 520A-D by extension) may be ordered such the telephone number included in the most private messages between the subject and another person is listed first (e.g., telephone number 510A).

In some embodiments, a user may interact with telephone numbers 510A-D via an I/O (e.g., I/O 270 of FIG. 2). The data structuring system may display a list of the private messages that included the telephone number 510A-D in response to the user's interaction.

FIG. 6 illustrates an example implementation of a conversation reconstruction data report 600. As shown in FIG. 6, data report 600 may include a list of private messages 610A-D. Private messages 610A-D may have been sent between a subject of a criminal investigation and another person (e.g., John Doe, entity 310A, and Jane Smith, entity 310B, both of FIG. 3). The data structuring system may generate display data report 600 by, for example, identifying private messages included in one or more object models (e.g., object model 300 of FIG. 3) stored in an object model database (e.g., object model database 250 of FIG. 2). The private messages may be identified based on the private messages with combinations of “TO” and “FROM” properties that include John Doe and Jane Smith.

Data report 600 allows users to interact with private messages 610A-D. For example, a user may select a private message 610A-D via an I/O. In the example illustrated in FIG. 6, private message 610A has been selected by the user. In response, the data structuring system may generate a detailed display 620 of selected private message 610A. For example, detailed display 620 may include the entire content of selected private message 610A, the “TO” and “FROM” properties of private message 610A, and the “DATE” and “TIME” properties of private message 610A.

FIGS. 7A and 7B illustrate an example implementation of a login information data report 600 and a mapped login information data report 730, respectively. As shown in FIG. 7A, data report 700 may include a timeline 710 across which login data 720 are distributed. Login data 720 may correspond to login events such as, for example, a subject of a criminal investigation logging into a social media platform. Each bar of login data 720 may represent the number of login events that occurred at certain points in time along timeline 710. Each bar of login data 720 may span a specified time duration. For example, each bar of login data 720 may cover a one-hour time interval, a 30-minute time interval, or any other time interval.

In some embodiments, data report 700 may be an interactive data report. For example, the data structuring system may be configured to receive input from a user corresponding to a selection of a subset of login data 620. The user may highlight a time interval of login data 720 along timeline 710. As shown in the example illustrated in FIG. 7A, a subset 730 has been selected.

A data report illustrating the subset 730 of login data 720 geographically mapped may be displayed in response to the data structuring system receiving the user's selection of subset 730. For example, mapped login information data report 740 illustrated in FIG. 7B may include the subset 730 of login data 720 superimposed over a map 750. A scale adjuster 770 may be used to zoom map 750 in and/or out so that more granularity can be obtained or more of subset 730 can be displayed at one time.

Data report 740 may illustrate the subject's locations 760 at the time of each login event included in the subset 730 of login data 720. In other words, locations 760 correspond to the subject's geographic location at the time the subject logged into the social media platform. In order to superimpose the subset 730 of login data 720 over map 750, the IP address properties associated with each login event may be traced by the data structuring system to obtain a set of geographic coordinates or other location data associated with the login event. Data structuring system may display the obtained location data as locations 760 over map 750.

It is to be understood that the example data reports illustrated in FIG. 5-7B are exemplary only and that other data reports are contemplated. Another example data report may include a picture matching report. For a picture matching report, a data structuring system may determine a identifier associated with a picture selected by a user and may use the identifier to identify all the social media platform profiles associated with the picture (e.g., that include the photo in a photo album, wall post, private message, etc.). Identifiers may include, for example, EXIF data, MD5 hash values, or other identifiers known in the art. The data structuring system may display the identified profiles as a graph, histogram, or any other format of data report.

Another data report may include a shared IP address data report. The shared IP address data report may include all the social media platform profiles associated with login events having the same IP address property. For example, a user may select an IP address associated with a subject of a criminal investigation logging into a social media platform. The data structuring system may determine all the social media platform profile logins using the same IP address, and display the identified profiles as a graph, histogram, or any other format of data report.

FIG. 8 depicts a flowchart of an example method 800, consistent with some embodiments and aspects of the present disclosure. Method 800 may be implemented, for example, for structuring data from unstructured electronic data files. The number and sequence of operations in FIG. 8 are provided for purposes of illustration and may be modified, enhance, substituted, or otherwise changed, in view of the present disclosure. In some embodiments, method 800 may be implemented as one or more computer programs executed by one or more processors. Moreover, in some embodiments, aspects of method 800 may be implemented by a data structuring system (e.g., data structuring system 130 having one or more processors executing one or more computer programs stored on a non-transitory computer readable medium) or a social media platform (e.g., social media platform 110, 120 having one or more processors executing one or more computer programs stored on a non-transitory computer readable medium). In some embodiments, method 800 may be implemented by a combination of a data importation system and a client device.

In some embodiments, example method 800 may include receiving an electronic data file at 810. For example, the data structuring system may receive legal process returns in the form of electronic data files from one or more social media platforms via a communications interface (e.g., communications interface 210 of FIG. 2). The legal process returns may be provided, for example, in response to a legal process such as a search warrant, national security letter, subpoena, etc., associated with a criminal investigation of a subject conducted by a law enforcement agency. The electronic data files may include any electronic file format and various types of structured and/or unstructured content. Example electronic data file formats include word processing files (.doc, .docx, .txt, .log, .rtf, etc.), spreadsheets (.xls, .xlsx, .ods, etc.), comma separated values (CSV) files, presentations, archived and compressed files (e.g., ZIP files, 7z files, cab files, RAR files, etc.), database files, PDF files, PUB files, image files, XML files, specialized tax and financial files (e.g., Open Financial Exchange and Interactive Financial Exchange files), tabulated data files and webpage files (e.g., HTML files). The content may include, for example, social media data associated with the subject of the criminal investigation as described above in reference to FIG. 1.

In some embodiments, example method 800 may include parsing the electronic data file to identify one or more objects included in the electronic data file at 820. For example, when the content included in the electronic data file received at 810 is unstructured content, the data structuring system may parse the unstructured data so that the data can be converted to a structured format. In some embodiments, the data structuring system includes a parser (e.g., parser 232 of FIG. 2) that parses the unstructured content using the parsing techniques described above in reference to FIG. 2. For example, the parser may identify words or strings of words in the received electronic data file and compare the identified words or strings of words to a selected set of object types defined in a database ontology (e.g., database ontology 240 of FIG. 2) to identify objects included in the electronic data file.

In some embodiments, example method 800 may include processing the unstructured content to identify one or more properties associated with the identified objects at 830. For example, the data structuring system may include a mapper (e.g., mapper 234 of FIG. 2) that assigns properties to the objects identified at 820. The objects, assigned object types, and assigned property types may be assigned to a structured object model (e.g., object model 300 of FIG. 3) of the electronic data file corresponding to the legal process return. In some embodiments, the object models may be stored in an object model database (e.g., object model database 250 of FIG. 2).

In some embodiments, example method 800 may include generating a data report at 840. For example, the data report may be generated by an object explorer of the data structuring system (e.g., object explorer 260 of FIG. 2). In some embodiments, the generated data report may be an interactive GUI (e.g., interactive GUI 400 of FIG. 4) that allows for free-form exploration and customization of the identified objects and properties. In some other embodiments, the generated data report may include any of the example data reports illustrated in FIGS. 5-7B and described above.

Embodiments of the present disclosure have been described herein with reference to numerous specific details that can vary from implementation to implementation. Certain adaptations and modifications of the described embodiments can be made. Other embodiments can be apparent to those skilled in the art from consideration of the specification and practice of the embodiments disclosed herein. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the present disclosure being indicated by the following claims. It is also intended that the sequence of steps shown in figures are only for illustrative purposes and are not intended to be limited to any particular sequence of steps. As such, it is appreciated that these steps can be performed in a different order while implementing the exemplary methods or processes disclosed herein.

Claims

1. A data structuring system for structuring data from electronic data files of a social media platform received in association with a legal process return, the system comprising: a memory device that stores a set of instructions;at least one processing device configured to execute the instructions to: receive an electronic data file, the electronic data file including unstructured content of the social media platform received in association with the legal process return;determine, based on scanning a header included in the electronic data file, that the electronic data file includes data received from the social media platform;identify, from a plurality of database ontologies corresponding to a plurality of social media platforms, a database ontology corresponding to the social media platform associated with the electronic data file, the database ontology defining known data objects and corresponding property types for unstructured data received from the social media platform;parse the unstructured content to identify a first string in the unstructured content;compare the first string to the known data objects defined in the database ontology corresponding to the social media platform, yielding a comparison;determine, based on the comparison, that the first string matches a first known data object defined in the database ontology corresponding to the social media platform, the first data object being of a first data object type that is associated with a first property type;in response to determining that the first string matches the first known data object, identify the first string as a first identified object in the unstructured content and assign the first object type to the first string;parse the electronic data file to identify a second string that follows the first string in the unstructured content;identify the second string as a first identified property of the first identified object in the unstructured content and assign the first property type to the second string; andgenerate a data report based on the first identified data object and the first identified property.

2. The system of claim 1, wherein the electronic data file associated with the legal process return is received in response to a legal process.

3. The system of claim 2, wherein the legal process includes at least one of a warrant, a national security letter, and a subpoena.

4. The system of claim 1, wherein the first identified data object is a private message, and the first identified property is an identifier included in the private message.

5. The system of claim 4, wherein the at least one processing device is configured to execute the instructions to: parse the identifier to normalize the identifier.

6. The system of claim 1, wherein the data report includes at least one of a list of histogramed telephone number data report, a conversation reconstructed from one or more private messages, a login information data report, a picture mapping data report, and a shared IP address data report.

7. A method of structuring data from unstructured electronic data files of a social media platform received in association with a legal process return, the method performed by at least one processing device and comprising: receiving an electronic data file, the electronic data file including unstructured content of the social media platform received in association with the legal process return;determining, based on scanning a header included in the electronic data file, that the electronic data file includes data received from the social media platform;identifying, from a plurality of database ontologies corresponding to a plurality of social media platforms, a database ontology corresponding to the social media platform associated with the electronic data file, the database ontology defining known data objects and corresponding property types for unstructured data received from the social media platform;parsing the unstructured content to identify a first string in the unstructured content;comparing the first string to the known data objects defined in the database ontology corresponding to the social media platform, yielding a comparison;determining, based on the comparison, that the first string matches a first known data object defined in the database ontology corresponding to the social media platform, the first data object being of a first data object type that is associated with a first property type;in response to determining that the first string matches the first known data object, identifying the first string as a first identified object in the unstructured content and assign the first object type to the first string;parse the electronic data file to identify a second string that follows the first string in the unstructured content;identifying the second string as a first identified property of the first identified object in the unstructured content and assign the first property type to the second string; andgenerating a data report based on the first identified data object and the first identified property.

8. The method of claim 7, wherein the electronic data file associated with the legal process return is received in response to a legal process.

9. The method of claim 8, wherein the legal process includes at least one of a warrant, a national security letter, and a subpoena.

10. The method of claim 7, wherein the data report is an interactive data report.

11. The method of claim 10, wherein the first identified data object is a private message, and the first identified property is an identifier included in the private message.

12. The method of claim 7, wherein the report includes at least one of a list of histogramed telephone number data report, a conversation reconstructed from one or more private messages, a login information data report, a picture mapping data report, and a shared IP address data report.

13. A non-transitory computer-readable medium storing a set of instructions that, when executed by one or more processors, cause the one or more processing devices to perform a method of structuring data from unstructured electronic data files of a social media platform received in association with a legal process return, the method comprising: receiving an electronic data file, the electronic data file including unstructured content of the social media platform received in association with the legal process return;determining, based on scanning a header included in the electronic data file, that the electronic data file includes data received from the social media platform;identifying, from a plurality of database ontologies corresponding to a plurality of social media platforms, a database ontology corresponding to the social media platform associated with the electronic data file, the database ontology defining known data objects and corresponding property types for unstructured data received from the social media platform;parsing the unstructured content to identify a first string in the unstructured content;comparing the first string to the known data objects defined in the database ontology corresponding to the social media platform, yielding a comparison;determining, based on the comparison, that the first string matches a first known data object defined in the database ontology corresponding to the social media platform, the first data object being of a first data object type that is associated with a first property type;in response to determining that the first string matches the first known data object, identifying the first string as a first identified object in the unstructured content and assign the first object type to the first string;parse the electronic data file to identify a second string that follows the first string in the unstructured content;identifying the second string as a first identified property of the first identified object in the unstructured content and assign the first property type to the second string; andgenerating a data report based on the first identified data object and the first identified property.