Sharing data with specific audiences

Abstract

Methods, apparatus, and non-transitory machine-readable media associated with sharing data with a specific audience are described. Examples may include: receiving first data at a processing resource; determining whether the first data includes a combination of bits associated with text or an image, or both; and comparing the combination of bits to second data stored on a memory resource. Examples may include: identifying one or more words or one or more images, or both, represented by the first data based on the comparison; and assigning first metadata representing a first security classification and a first confidence level and second metadata representing a second security classification and a second confidence level to the first data. Examples may include transmitting an output including the first data or third data, the third data including a combination of bits modified relative to the combination of bits of the first data.

Description

Technical Field

[0001] The present disclosure generally relates to devices, non-transitory machine-readable media, and methods associated with sharing data with a specific audience. Background Art

[0002] Data security involves protecting data (e.g., data in a database) from destructive forces and from undesirable actions by unauthorized users, such as cyberattacks or data exfiltration. Data security can include protecting data from unauthorized access and data corruption throughout its lifecycle. Data security includes data encryption, hashing, tokenization, and / or key management practices that protect data across different applications and platforms.

[0003] In some cases, it may be necessary to share secure or non-secure data. Data sharing can include sharing data between two or more parties (e.g., individuals, organizations, or a combination thereof). Data can be shared from one party to another or between parties. When the data is secure, authorization may be required before sharing. Summary of the Invention

[0004] In one aspect, the present application provides a method comprising: receiving first data at a processing resource; determining at the processing resource whether the first data includes a combination of bits associated with text or an image or both; comparing the combination of bits with second data stored on a memory resource or a storage device coupled to the processing resource via a network; identifying one or more words represented by the first data or one or more images represented by the first data or both based at least in part on the comparison; assigning first metadata representing a first security classification and a first confidence level based at least in part on the identified one or more words to the first data, and second metadata representing a second security classification and a second confidence level based at least in part on the identified one or more images; and transmitting an output comprising the first data or third data to a set of users via the network and based at least in part on the first or second security classification and the first or second confidence level, the third data comprising a combination of bits modified relative to the combination of bits of the first data.

[0005] In another aspect, the present application further provides a non-transitory machine-readable medium comprising a processing resource in communication with a memory resource having instructions, the instructions being executable to: receive a request to share first data received at the processing resource with a specific audience; use artificial intelligence (AI) to determine a first portion of the first data that can be shared with the specific audience, a second portion of the first data that cannot be shared with the specific audience, a third portion of the first data having an undetermined sharing status, or a combination thereof based on a comparison of the request with second data stored on a memory resource communicatively coupled to the medium; request approval from an administrator to share the third portion with the specific audience; write data associated with the request, the first portion, the second portion, and the third portion to the memory resource; watermark the first portion and the third portion and share the first portion and the third portion with the specific audience in response to the third portion being approved for sharing; and watermark the first portion and share the first portion with the specific audience in response to the third portion being approved for sharing. BRIEF DESCRIPTION OF THE DRAWINGS

[0006] Figure 1 is a flowchart representing an example method for sharing data with a specific audience according to several embodiments of the present disclosure.

[0007] Figure 2 is another flow chart representing an example method for sharing data with a specific audience according to several embodiments of the present disclosure.

[0008] Figure 3 is a functional diagram representing a processing resource in communication with a memory resource having instructions written thereon, according to several embodiments of the present disclosure.

[0009] Figure 4 is another functional diagram representing a processing resource in communication with a memory resource having instructions written thereon, according to several embodiments of the present disclosure.

[0010] Figure 5 is yet another flow chart representing an example method for sharing data with a specific audience according to several embodiments of the present disclosure.

[0011] Figure 6 is another flow chart representing an example method for sharing data with a specific audience according to several embodiments of the present disclosure. DETAILED DESCRIPTION

[0012] Devices, machine-readable media, and methods related to sharing data with specific audiences are described. An individual or organization may need to share specific data with different users or organizations, but sharing may require authorization by one or more administrators. This process may include obtaining authorization from the administrator each time data sharing is desired regarding what portion of the data (if any) may be shared and with whom the data may be shared. This process may be time-consuming and may result in inconsistent sharing depending on who approved / disapproved the sharing of specific data and when the approval / disapproval was received.

[0013] Examples of the present disclosure can improve the efficiency of data sharing. Data sharing according to the present disclosure can shorten the time spent in the authorization process by utilizing a data sharing database, against which the data desired to be shared can be compared. Data sharing decisions can be based on the comparison, which can improve the efficiency and consistency of data sharing approvals / rejections. Examples of the present disclosure can utilize artificial intelligence (AI) to improve data sharing decisions using historical and other data associated with current and previous data sharing requests. As used herein, AI includes controllers, computing devices, or other systems that perform tasks that typically require human intelligence. For example, as further described herein, a sharing tool can organize data and make data sharing decisions that typically require human intelligence. AI can include the use of one or more machine learning models.

[0014] Examples of the present disclosure may include a method for sharing data with a specific audience, comprising: receiving first data at a processing resource; determining at the processing resource whether the first data includes a combination of bits associated with text or an image, or both; and comparing the combination of bits with second data stored on a memory resource or storage device coupled to the processing resource via a network. Examples may include identifying one or more words represented by the first data or one or more images represented by the first data, or both, based at least in part on the comparison, and the method may further include assigning first metadata to the first data representing a first security classification and a first confidence level based at least in part on the identified one or more words, and second metadata representing a second security classification and a second confidence level based at least in part on the identified one or more images. Examples may include transmitting an output including the first data or third data to a collection of users via the network and based at least in part on the first or second security classification and the first or second confidence level, the third data including a combination of bits modified relative to the combination of bits of the first data.

[0015] Other examples of the present disclosure may include a non-transitory machine-readable medium comprising a processing resource in communication with a memory resource having instructions, the instructions being executable to receive a request to share first data received at the processing resource with a specific audience. The example may include instructions executable to: determine, using AI, based on a comparison of the request with second data stored on the memory resource, a first portion of the first data that can be shared with the specific audience, a second portion of the first data that cannot be shared with the specific audience, a third portion of the first data that has an undetermined sharing status, or a combination thereof. The instructions are executable to request approval from an administrator to share the third portion with the specific audience; write first data associated with the request, the first portion, the second portion, and the third portion to the memory resource; watermark the first portion and the third portion and share the first portion and the third portion with the specific audience in response to the third portion being approved for sharing; and watermark the first portion and share the first portion with the specific audience in response to the third portion being approved for sharing.

[0016] Still other examples of the present disclosure may include the same or different non-transitory machine-readable media, comprising a processing resource in communication with a memory resource having instructions, the instructions executable to: use AI to classify received first data having a combination of bits associated with text or an image, or both; use AI to classify the first data into one of a plurality of predetermined security levels based on an analysis of the classified first data; and use AI to assign a confidence level to the security level classification. The instructions executable to compare the combination of bits with second data stored on the memory resource, and determine a sharing status of the first data based on the comparison, the security level classification, and the confidence level, the sharing status including determining whether the first data can be shared with a specific audience, cannot be shared with the specific audience, or whether it is undetermined whether the first data can be shared with the specific audience.

[0017] In the following detailed description of the present disclosure, reference is made to the accompanying drawings which form a part hereof and in which is shown by way of illustration the manner in which one or more embodiments of the present disclosure may be practiced. These embodiments are described in sufficient detail to enable one skilled in the art to practice the embodiments of the present disclosure, and it is to be understood that other embodiments may be utilized and process, electrical, and structural changes may be made without departing from the scope of the present disclosure.

[0018] It should also be understood that the terms used herein are for the purpose of describing specific embodiments only and are not intended to be limiting. As used herein, unless the context clearly dictates otherwise, the singular forms "a / an" and "the" may include both singular and plural referents. In addition, "several," "at least one," and "one or more" (e.g., several memory devices) may refer to one or more memory devices, while "multiple" is intended to refer to more than one of such things. In addition, throughout this application, the words "may" and "can" are used in a permissive sense (i.e., it is possible, can), rather than in a mandatory sense (i.e., must). The term "including" and its derivatives refer to "including (but not limited to)." Depending on the context, the term "coupled / coupling" means physically connected directly or indirectly or used to access and move (transmit) commands and / or data.

[0019] The figures herein follow a numbering convention in which the first one or more digits correspond to the figure number and the remaining digits identify an element or component in the figure. Similar numerals may be used to identify similar elements or components between different figures. For example, 106 may represent Figure 1 Component "06" in the , and similar components can be found in Figure 2 denoted as 206. As will be appreciated, elements shown in the various embodiments herein may be added, exchanged, and / or removed to provide several additional embodiments of the present disclosure. Additionally, the proportions and / or relative scales of the elements provided in the figures are intended to illustrate certain embodiments of the present disclosure and should not be viewed in a limiting sense.

[0020] Sharing data may include authorization. Instances of the present disclosure may include user requests to share data (e.g., specific inputs) with specific audiences. For example, an individual or team at organization A may need to share a workflow with an individual or team at organization B. Instances of the present disclosure allow individuals or teams at organization A to upload data that may contain sensitive and / or secure data to a sharing tool to determine whether the data can be shared with a specific audience (e.g., organization B), cannot be shared with a specific audience, or require approval from an administrator. In some instances, the sharing tool may be a self-learning AI sharing tool, such that the sharing tool learns and updates itself when a data sharing request is made. In some instances, the sharing tool includes a machine learning model.

[0021] Figure 1 is a flowchart representing an example method for sharing data with a specific audience according to several embodiments of the present disclosure. Figure 1 , the data that is expected to be shared is called "specific input" so that Figure 1The data desired to be shared or "specific input" may include documents, data, images, videos, or other data that may require authorization before being shared. For example, the data desired to be shared or "specific input" may include a document that an employee of the first company wishes to share with an employee of the contracted second company. For example, the document cannot be shared without authorization.

[0022] At 102, a configured category is selected. For example, a configured category may include a category or partition into which a particular input may fall. This may be based on a department within an organization, a project to which the particular input belongs, data associated with the particular input (e.g., type of memory used, budget, yield, subject classification, etc.). The user may select an option provided from a predetermined list, or may be able to create a new configured category. Additional data may be included with a particular input (e.g., images, definitions, supporting documentation, reason for the request, etc.), and at 104, the particular input may be uploaded to the sharing tool 106.

[0023] In some instances, the sharing tool 106 may include a processing resource in communication with a memory resource that utilizes an AI to organize specific inputs into shareable, non-shareable, and ambiguous categories. In other words, the sharing tool 106 creates a sharing state for a specific input, a portion of a specific input, or both (e.g., a "shareable" sharing state, a "non-shareable" sharing state, an "undetermined" sharing state, etc.). The sharing tool 106 (and associated AI (e.g., including a machine learning model) may be trained using a training data set. For example, the training data set may include a collection of examples for fitting parameters of the AI. For example, the training data set may include data associated with shareable image data, text data, both image and text data, and associated shareability. In some instances, the sharing tool 106 may also be trained using new input data (e.g., specific inputs, etc.).

[0024] Within the shared tool 106, the method may include AI identifying file types, such as documents (e.g., PDFs, word processors, etc.), images, paragraphs from a specific file format, and / or circuit board files (e.g., circuitry, connections in a memory type (e.g., NAND), etc.). At 108, the specific input may be classified as text or an image, or both. For example, a determination may be made at the processing resource as to whether the specific input includes a combination of bits associated with text or an image, or both. The classification of the specific input may include, for example, text searching, image recognition, a "looks like" tool, file parsing, binary decoding, or a combination thereof, etc.

[0025] At 110, the method may include analyzing the specific input. For example, intellectual property within the specific input may be identified (e.g., using text recognition, image recognition, etc.). Identifying intellectual property may allow identification of data associated with the specific input that can be removed from shared data, or may prevent sharing altogether. Intellectual property may include creative ideas of the mind, such as inventions, literary and artistic works, designs, and symbols, names, and images used in trade. As used herein, intellectual property may be data that an individual or organization does not wish to share or may choose to share in limited circumstances (e.g., under a nondisclosure agreement).

[0026] At 112, historical data associated with the particular input is analyzed. Utilizing AI, a determination may be made as to whether data having the same components having the same purpose as the particular input has been previously shared with a particular audience. For example, a memory resource or storage device (e.g., a database) coupled to the processing resource via a network may be maintained that includes historical data from previous requests to share data, and a comparison may be made between the particular input and the memory resource or storage device. Matches during the comparison, including matches where the particular input has been previously shared with the particular audience or matches where the particular input has definitely not been previously shared with the particular audience, may affect the sharing determination. For example, if the particular input has been shared with the particular audience historically, authorization may be given (e.g., via sharing tool 106) to share the particular input again. In some examples, a match may include a match of a portion of the particular input, such as particular text, an image, or an image fragment.

[0027] At 114, based on the input classification, analysis of the particular input, and analysis of historical data, the particular input may be classified as one of a plurality of predetermined security levels (e.g., "Highly Confidential," "Confidential," "Internal Only," "Public," etc.). For example, metadata representing the security classification may be assigned to the particular input. This classification may be performed automatically, such that the classification is performed with limited or no user input and / or with limited or no prompting. For example, if particular text in the particular input matches text in a memory resource or storage device that has been previously classified as highly confidential, the particular input, or at least a portion of the particular input, may be given the same security classification.

[0028] In some examples, a confidence level can be assigned to a security classification. For example, metadata representing a confidence level can be assigned to a particular input. In some examples, if sharing tool 106 has a high confidence level for the security classification (e.g., exact text matches, image segment matches, etc. for all text and images in the particular input), the confidence level can be high (e.g., closer to 100%), while if sharing tool 106 has a low confidence level for the security classification (e.g., some text matches, some image segment matches, etc. for text and images in the particular input), the confidence level can be low (e.g., further away from 100%). The confidence level can be based on, for example, a training dataset provided to sharing tool 106. The training dataset can include a collection of examples used to fit parameters of sharing tool 106 (e.g., for use by the sharing tool's AI (e.g., and / or associated machine learning algorithms)). Sharing tool 106 can be trained on the training dataset to make determinations regarding confidence levels. Sharing tool 106 can be updated based on the particular input and / or other inputs.

[0029] The method may include requesting approval from an administrator or group of administrators if the sharing status is undetermined at 116. For example, if a match between a particular input and a memory resource or storage device cannot be made, if only a partial match can be made (e.g., the particular input has been shared with a different audience), or if the confidence associated with the security classification is low, one or more administrators may be prompted to review the request. For example, a group of administrators may be identified to poll for approval or rejection of the request. The group may include supervisors, department heads, lead engineers, project managers, individuals familiar with the project, etc. Once the request is submitted, the group's review progress may be tracked, and the group members, their decisions, and their explanations of their decisions may be added to the memory resource or storage device, and the sharing tool 106 and associated AI may be updated to reflect this data for self-learning purposes and / or future review. The decision may include sharing the particular input with a particular audience, sharing a portion of the particular input with a particular audience (e.g., deleting certain portions), or denying sharing of the particular input with a particular audience.

[0030] The method may also include, at 116, sending a notification to an administrator or group of administrators if the sharing status is approved, converting the specific input to an uneditable format, and watermarking the specific input. For example, if the sharing tool 106 determines that the specific input can be shared with a specific audience, the sharing tool 106, the memory resource or storage device, and the AI ​​may be updated, and the administrator (e.g., someone in the associated group) may be notified. The specific input is protected in an uneditable format and watermarked for security purposes and prepared for sharing. In some instances, even if the sharing tool 106 has approved sharing, the administrator may still be able to prevent the sharing of the specific input with a specific audience.

[0031] At 118, the sender of the request to share the specific input may be notified of the sharing tool 106 results. For example, he or she may be notified whether the sharing was approved, denied, or the request was sent to an administrator for review. If the sharing tool 106 or group approves, the sender of the request may share the specific input with the specific audience.

[0032] The method may include updating the sharing tool 106 and associated AI and memory resources or storage based on data associated with the specific input, as discussed herein, at 120. For example, details of the request, the department and project associated with the request, the administrator associated with the request, the results of the request, the specific audience, etc. may be saved in the memory resources or storage, and the sharing tool 106 may self-learn to update and improve the accuracy and efficiency of data sharing decisions.

[0033] At 122, feedback can be requested regarding the sharing tool 106. The sender of the request can be prompted to take a survey or leave feedback regarding the usability, visuals, usefulness, and performance of the sharing tool 106. Based on the received feedback, the sharing tool 106 can be adjusted. For example, based on the feedback, the sharing tool 106 can be adjusted to improve the user interface, accessibility, visuals, etc.

[0034] Figure 2 is another flow chart representing an example method for sharing data with a specific audience according to several embodiments of the present disclosure. At 228, a request is made to share first data (e.g., a specific input) with a specific audience. For example, a processing resource may receive the first data and may determine whether the first data includes a combination of bits associated with text or an image or both. That is, the first data is classified as text or text data at 230, an image or image data at 232, or one of the two (e.g., a combination of text and image data) at 234. The first data may be uploaded to a sharing tool 206, which may include or be communicatively coupled to a processing resource, which may utilize AI to determine whether the first data may be shared with the specific audience, may not be shared with the specific audience, or whether the sharing status is undetermined and authorizes one or more administrators to conduct further analysis to make a sharing determination.

[0035] In some examples, the combination of bits may be compared to second data stored on a memory resource or storage device (e.g., a database) coupled to the processing resource via a network, and based at least in part on the comparison, one or more words represented by the first data or one or more images represented by the first data, or both. Figure 2 Each of the text, image, and both text and image instances are discussed separately.

[0036] For example, if the first data is determined to be text, specific words in the text may be detected and / or identified at 236, for example, using a text recognition tool. The words may be organized and a determination of their content may be made. For example, the sharing tool 206 may have a pool of specific words that will not be shared, a pool of words that may be shared, or a combination thereof. Using these pools and the specific words, metadata representing a security classification may be assigned to each of the words at 238 (e.g., based at least in part on the identified words). For example, based on the pools, some words may be classified as "highly confidential" (e.g., trade secret words, organization-specific words, etc.), while other words may be classified as "confidential," "internal use only," or "public." Different levels of security are possible, and the security levels are not limited to the four levels described above.

[0037] At 240, metadata representing confidence levels may be assigned to different security classifications (e.g., based at least in part on the recognized text). As used herein, confidence levels include a degree of confidence in the security classification performed by sharing tool 206. For example, a 0% confidence level means there is absolutely no confidence that the results would be the same if the analysis were repeated (or if performed by a human). A 100% confidence level means there is absolutely no doubt that the results would be the same if the analysis were repeated (or if performed by a human). In other words, the higher the confidence level, the more confidence there is in the security classification. For example, a particular text that matches the "unshared" text in the pool may have a higher confidence level associated with its security classification than a particular text that matches neither the "unshared" text nor the "shared" text in the pool. The first data or portion of the first data may be labeled based on its security classification and / or its confidence level. For example, a high confidence level may be assigned a green flag label, while a low confidence level may be assigned a red flag label. A highly confidential security classification may be assigned a purple flag, while a public security classification may be assigned an orange flag. Labeling is not limited to these categories or colors. In some instances, security level classifications may be labeled separately from confidence levels. Labeling can be used by the sharing tool 206 or administrators to improve efficiency when making sharing decisions.

[0038] At 226-1, the first data may be compared to a memory resource or storage device (e.g., a database) to determine whether input having the same components has been previously shared with a specific audience. For example, if specific text from the first data matches text in the memory resource or storage device that has been previously shared with the specific audience, a match is made. In some instances, only some specific text may match the memory resource or storage device, which may result in the deletion of unmatched specific text. For example, if the specific text does not match text in the memory resource or storage device that has been previously shared with the specific audience, or if the specific text does match text in the memory resource or storage device but those text has not been previously shared with the specific audience, no match is made. In some instances, the specific text may be matched to text in the memory resource or storage device that was previously not allowed to be shared with the specific audience. Based on these comparisons, the sharing tool 206 may output data associated with the first data at 224, as will be discussed further herein.

[0039] If the first data is considered to be an image, one or more images may be reviewed and / or adjusted at 246 . For example, the images may be placed in the same orientation so that they are not tilted, rotated, etc. differently from each other. Each image may be segmented, and each segment may be analyzed at 244 to determine the content. For example, the sharing tool 206 may have a pool of images that are not to be shared, a pool of images that are to be shared, or a combination thereof. Using these pools and the particular image, metadata representing a security classification may be assigned to each of the image segments at 242 (e.g., based at least in part on the identified image). For example, based on the pool, some image segments may be classified as "highly confidential" (e.g., trade secret text, organization-specific text, etc.), while other text may be classified as "confidential," "internal use only," or "public." Different security levels are possible, and the security levels are not limited to the four levels described above.

[0040] Also at 242, metadata representing confidence levels can be assigned to different security classifications (e.g., based at least in part on the identified image). For example, a particular image segment that matches an "unshared" image in a pool (e.g., a pool within a processing resource or storage device) can have a higher confidence level associated with its security classification than a particular image segment that matches neither the word "unshared" nor the word "shared" in the pool. In some examples, the image segment security classification and confidence levels can be analyzed together with respect to the overall image security classification and associated confidence levels. In some examples, the security level classification and / or confidence levels can be annotated for ease of review.

[0041] In a non-limiting example, an image is divided into segments, and each segment is identified. For example, an image may be divided into segments 1, 2, 3, and 4. Segments 1 and 2 may be matched to images in a pool, or image recognition tools may be used to identify segments 1 and 2 as shareable images (e.g., "Public"). Segment 3 may be determined to be non-shareable image data (e.g., "Highly Confidential"), and segment 4 may be undetermined. Based on the security classification and associated confidence level, a sharing determination may be made.

[0042] At 226-2, the first data may be compared to a memory resource or storage device (e.g., a database) to determine whether input having the same components has been previously shared with a specific audience. For example, if a specific image segment from the first data matches an image segment in the memory resource or storage device that has been previously shared with the specific audience, a match is made. In some instances, only some specific image segments may match the memory resource or storage device, which may result in the deletion of unmatched specific image segments. For example, if a specific image segment does not match an image segment in the memory resource or storage device that has been previously shared with the specific audience, or if a specific image segment does match an image segment in the memory resource or storage device but those image segments have not been previously shared with the specific audience, no match is made. In some instances, the specific image segment may be matched to an image segment in the memory resource or storage device that was previously not allowed to be shared with the specific audience. Based on these comparisons, the sharing tool 206 may output data associated with the first data at 224, as will be discussed further herein.

[0043] If the first data is considered to be a combination of text and one or more images, the sharing tool 206 can detect an area (e.g., a rectangular area) within the first data that may be text at 248 and separate the text from the image at 250. At 252, specific words in the text can be detected and / or recognized as previously discussed herein, and at 254, the one or more images can be reviewed and / or adjusted and segmented as previously discussed herein.

[0044] At 256, metadata representing a security classification and confidence level may be assigned to specific text, image segments, and images as previously described herein (e.g., based at least in part on the identified text, image, or both). In some instances, the security level classification and / or confidence level may be annotated for ease of review. At 226-3, the first data may be compared to a memory resource or storage device (e.g., a database) to determine whether input having the same components has been previously shared with a particular audience. For example, as previously discussed herein, text may be compared to text data in a memory resource or storage device, and one or more images may be compared to image data in a memory resource or storage device. Based on these comparisons, the sharing tool 206 may output data associated with the first data at 224.

[0045] For example, at 224, the sharing tool 206 may output the results of the data sharing analysis. The results of the analysis may include confidence levels, and actions may be taken based on the results and those associated confidence levels. In other words, the output may include the first data or third data including a modified bit pattern of the first data (e.g., deleted data, etc.), and may be transmitted to a collection of users via a network based at least in part on the security classification and confidence levels. For example, if a determination is made that the first data may be shared with a specific audience, the first data may be converted to a non-editable format and watermarked. However, if the confidence level is low, further analysis by an administrator may be desired.

[0046] If a determination is made that the sharing status of the first data is undetermined, an approval request may be sent to one or more administrators along with any confidence level of the determination. The requesting party may be notified of the results from the sharing tool 206 or the administrator, along with other associated data (e.g., security level classification, confidence level, etc.), regardless of the determination.

[0047] Figure 3 is a functional diagram showing a processing resource 362 in communication with a memory resource 360 ​​having instructions 364, 366, 368, 370, 372 written thereon, according to several embodiments of the present disclosure. In some examples, the processing resource 362 and the memory resource 360 ​​include, for example, Figure 1 and 2 Sharing tools such as sharing tools 106 or 206 shown in .

[0048] Figure 3The system shown can be a server or a computing device (among others) and can include processing resources 362. The system can further include memory resources 360 (e.g., non-transitory MRMs) on which instructions, such as instructions 364, 366, 368, 370, and 372, can be stored. Although the following description refers to processing resources and memory resources, the description can also be applied to systems having multiple processing resources and multiple memory resources. In such examples, instructions can be distributed (e.g., stored) across the multiple memory resources, and instructions can be distributed (e.g., executed by) the multiple processing resources.

[0049] Memory resource 360 ​​may be an electronic, magnetic, optical, or other physical storage device that stores executable instructions. Thus, memory resource 360 ​​may be, for example, nonvolatile or volatile memory. For example, nonvolatile memory provides persistent data by retaining written data when power is not supplied, and nonvolatile memory types may include NAND flash memory, NOR flash memory, read-only memory (ROM), electrically erasable programmable ROM (EEPROM), erasable programmable ROM (EPROM), and storage-class memory (SCM) that may include resistance variable memory, such as phase-change random access memory (PCRAM), three-dimensional cross-point memory, resistive random access memory (RRAM), ferroelectric random access memory (FeRAM), magnetoresistive random access memory (MRAM), and programmable conductive memory, among other types of memory. Volatile memory may require power to maintain its data and may include random access memory (RAM), dynamic random access memory (DRAM), and static random access memory (SRAM), among others.

[0050] In some examples, memory resource 360 ​​is a non-transitory MRM, including random access memory (RAM), electrically erasable programmable ROM (EEPROM), a storage drive, an optical disk, etc. Memory resource 360 ​​may be located within a controller and / or computing device. In this example, executable instructions 364, 366, 368, 370, 372 may be "installed" on the device. Additionally and / or alternatively, memory resource 360 ​​may be a portable external or remote storage medium, for example, allowing the system to download instructions 364, 366, 368, 370, 372 from the portable / external / remote storage medium. In this case, the executable instructions may be part of an "installation package." As described herein, memory resource 360 ​​may be encoded with executable instructions for performing dehumidification actions in the environment of the electrical component.

[0051] Instructions 364, when executed by a processing resource such as processing resource 362, may include instructions for using AI to classify received first data (e.g., a specific input received at processing resource 362) as having a combination of bits associated with text, images, or both. For example, an image recognition tool may be used to determine whether the first data includes one or more images. If one or more images are detected, a check may be made for text tracking using, for example, a text recognition tool and / or a tool for separating images and text. If no images are detected, the first data may be determined to include text.

[0052] Instructions 366, when executed by a processing resource such as processing resource 362, may include instructions for using AI to classify the first data into one of a plurality of predetermined security levels based on the analyzed content of the classified first data. For example, text, images, or a combination thereof may be analyzed for particular text and / or image segments. Based on the content of these text and / or image segments, a determination may be made regarding a security level classification, and associated metadata may be assigned to the first data. For example, AI may be used to determine whether received text contains trade secret text that should be placed in a "highly confidential" security level classification, and / or whether the text contains text that should receive a "public" or other security level classification. Similarly, AI may be used to determine a security level classification for a received segment of one or more images.

[0053] Instructions 368, when executed by a processing resource such as processing resource 362, may include instructions for using AI to assign a confidence level to a security level classification. In such an example, metadata associated with the confidence level may be assigned to the first data. The confidence level may include a confidence level for the security level classification, where a higher confidence level indicates greater trust in the security level classification than a lower confidence level. For example, a particular text or image segment that matches "unshared" text or image segments in a pool of text or image segments used for comparison may have a higher confidence level associated with its security classification than a particular text or image segment that matches neither "unshared" nor "shared" text or image segments in the pool.

[0054] Instructions 370, when executed by a processing resource such as processing resource 362, may include instructions to compare the combination of bits with second data stored on a memory resource 360 ​​or storage device (e.g., a database) coupled to the processing resource via a network. Memory resource 360 ​​or storage device may include data associated with previous requests to share input, including the type of input, with whom the input was shared, who approved the sharing, when and how the input was shared, why the input was shared, the data file type of the input, and inputs that were blocked from sharing, etc. The first data may be compared with the data in memory resource 360 ​​or storage device to determine whether data having the same components, the same sharing purpose, and the same proposed audience as the first data has been previously shared, has been previously blocked from sharing, or neither.

[0055] Instructions 372, when executed by a processing resource such as processing resource 362, may include instructions for determining a sharing status of the first data based on the comparison, the security level classification, and the confidence level, wherein the sharing status includes determining whether the first data can be shared with a specific audience, cannot be shared with a specific audience, or whether it is not determined whether the first data can be shared with a specific audience.

[0056] For example, if a match is made between the first data and the memory resource 360 ​​or storage device indicating that the first data is shareable and the confidence level in the security level classification is high, the sharing status can be "shareable," indicating that the first data can be shared with a specific audience. If a match is made between the first data and the memory resource 360 ​​or storage device indicating that the first data is not shareable and the confidence level in the security level classification is high, the sharing status can be "unshareable," indicating that the first data is not shareable with a specific audience. If the confidence level is low in either case, the first data can be sent to one or more administrators for further review with a sharing status of "undetermined." Similarly, if no match is made between the first data and the memory resource 360 ​​or storage device, the sharing status can be "undetermined," and the first data can be sent to one or more administrators for further review.

[0057] In some examples, the instructions may be executable to, in response to the sharing status indicating that the first data is shareable, notify the sender of the request that the first data is shareable with a specific audience. For example, if the first data matches data previously shared with a specific audience, the sender may be notified that they may begin sharing. In some examples, the sender may be instructed as to which portions of the first input may be shared with the specific audience. For example, the sender may be instructed to delete highly confidential portions.

[0058] The instructions are executable to notify the sender of the request that the first data cannot be shared with the specific audience in response to the sharing status indicating that the first data cannot be shared. For example, if the first data matches data that has previously been denied sharing with the specific audience, the sender may be notified that it cannot begin sharing.

[0059] In some instances, the instructions may be executable to request approval from an administrator to share the first data with a specific audience in response to the undetermined sharing status, and notify the sender of the result of the approval request. For example, if there is no match with the first data (approval or rejection), the one or more administrators may be prompted to review the request. If the one or more administrators indicate that the first data can be shared with the specific audience, the sender may be notified that it can begin sharing. In some examples, the sender may be given instructions regarding which parts of the first input can be shared with the specific audience. For example, the sender may be instructed to delete highly confidential portions. If the one or more administrators indicate that the first data cannot be shared with the specific audience, the sender may be notified that it cannot share the first data with the specific audience.

[0060] In some examples, in response to being classified as text, instructions 364 may be executed to detect a specific word or group of words in the text, classify the specific word or group of words into a corresponding one of the plurality of security levels using AI, and assign a confidence level (e.g., metadata associated with the first data) to each of the assigned security levels using AI. The specific word or group of words may be compared with words in the memory resource 360 ​​or storage device to determine the sharing status of the first data.

[0061] In some examples, in response to being classified as an image, instructions 364 may be executed to adjust the orientation of the image, segment the image into a plurality of segments, classify each of the plurality of segments into a respective one of the plurality of security levels using AI, and assign a confidence level (e.g., metadata associated with the first data) to each of the assigned security levels using AI. The image segments may be compared with images in the memory resource 360 ​​or storage device to determine the sharing status of the first data.

[0062] In some examples, in response to being classified as a combination of text and one or more images, instructions 364 may be executed to divide the first data into text and images, detect specific words or groups of words in the text, adjust the orientation of the image, and segment the image into a plurality of segments. The specific words or groups of words may be classified using AI, and each of the plurality of segments may be classified into a corresponding one of the plurality of security levels, and a confidence level (e.g., metadata associated with the first data) may be assigned using AI. The specific words or groups of words and the image segments may be compared with the words and images in the memory resource 360 ​​or storage device to determine the sharing status of the first data.

[0063] Figure 4 is another functional diagram showing a processing resource 476 in communication with a memory resource 474 having instructions 478, 480, 482, 484, 486 written thereon, according to several embodiments of the present disclosure. In some examples, the processing resource 476 and the memory resource 474 may be similar to the respective processors as shown with respect to FIG. Figure 3 The processing resources 362 and memory resources 360 are described. In some examples, the processing resources 476 and memory resources 474 include, for example, Figure 1 and 2 Sharing tools such as sharing tools 106 or 206 shown in .

[0064] Instructions 478, when executed by a processing resource such as processing resource 474, may include instructions for receiving a request to share first data received at processing resource 474 with a specific audience. For example, the request may be uploaded to a sharing tool for data sharing decisions. In some instances, a determination may be made as to whether the first data includes a combination of bits associated with text or an image, or both (e.g., a combination of text and an image). For example, text recognition, image recognition, and other tools may be used to determine the data type (e.g., an image, text, or both). In some examples, AI may be used to identify industry standard file types associated with the first data, such as documents (e.g., pdf, word processing programs, etc.), images, paragraphs from a specific file format, and / or circuit board files (e.g., connections in a circuit system, a memory type (e.g., NAND), etc.).

[0065] In some examples, security level categories (e.g., "Highly Confidential," "Confidential," "Internal Only," "Public," etc.) can be assigned to the first, second, and third portions, and confidence levels can be assigned to the security level classifications using AI (e.g., together with associated metadata assigned to the first data). For example, if, based on the training data, the first data matches data in the pool that was previously classified as "Public" with a lower confidence level, the confidence level in the security level classification of the first data can be lower than if the first data matches data in the pool that was previously classified as "Public" with a higher confidence level. The first, second, and third portions can be labeled with identification tags based on their associated security level categories, associated confidence levels, or both.

[0066] The instructions 480, when executed by a processing resource such as the processing resource 474, may include instructions for determining, using the AI, a first portion of first data that is shareable with a particular audience, a second portion of first data that is not shareable with the particular audience, a third portion of first data that has an undetermined sharing status, or a combination thereof based on a comparison of the request with second data stored on a memory resource 474 or storage device (e.g., a database) that is coupled to the processing resource 476 via a network that is communicatively coupled to the medium. For example, all of the first data may be shareable with the particular audience (e.g., if all of the first data exactly matches a previous request in the memory resource 474 or storage device that is shareable with the particular audience), or none of the first data may be shareable with the particular audience (e.g., if all of the first data exactly matches a previous request in the memory resource 474 or storage device that is deemed not shareable with the particular audience). In some instances, for example, if some of the first data exactly matches a previous request in a memory resource 474 or storage device that can be shared with a particular audience, but some portions do not match (e.g., a "highly confidential" portion that can be deleted), then a portion of the first data can be shared with the particular audience as long as the other portions are deleted.

[0067] Instructions 482, when executed by a processing resource such as processing resource 474, may include instructions for requesting approval from an administrator to share the third portion with the specific audience. If a determination cannot be made using memory resource 474 or storage, for example, if a shareable match or a non-shareable match cannot be made, one or more administrators (e.g., a group of administrators) may be prompted to make a decision based on their knowledge of the subject matter, the first data, the specific audience, and historical methods. The one or more administrators may decide that the first data may be shared with the specific audience, may not be shared with the specific audience, or may decide that a portion of the first data may be shared with the specific audience (e.g., portions must be deleted before sharing).

[0068] The instructions 484, when executed by a processing resource such as the processing resource 474, may include instructions to write data associated with the request, the first portion, the second portion, and the third portion (e.g., first data) in the memory resource 474 or the storage device. For example, the data associated with the request, the first portion, the second portion, and the third portion in the memory resource 474 or the storage device may include whether the data has been shared, whether the data is currently being shared, whether the data is planned to be shared, with whom the data was or is planned to be shared, who requested sharing of the data, an administrator, and keywords associated with the data, or a combination thereof.

[0069] In some examples, the written data associated with the request, the first portion, the second portion, and the third portion in the memory resource 474 or storage device may be used to update the memory resource 474 or storage device. The updated memory resource 474 or storage device, along with the updates to the AI, may allow for self-learning and improved accuracy, efficiency, and consistency in the data sharing decision-making process.

[0070] Instructions 486, when executed by a processing resource such as processing resource 474, may include instructions for watermarking the first and third parts and sharing the first and third parts with a specific audience in response to the third part being approved for sharing, and watermarking the first part and sharing the first part with a specific audience in response to the third part being approved for sharing.

[0071] In some examples, performance input regarding the medium may be received from the party sending the request, a specific audience, or a combination thereof and implemented into the medium. For example, the party may receive a request (e.g., a survey) to rate their experience with the sharing tool after completing the sharing request process.

[0072] Figure 5 is another flow chart illustrating an example method 590 for sharing data with a specific audience according to several embodiments of the present disclosure. The method 590 may be performed by, for example, Figure 3 and 4 The system described above is executed by the system. Figure 1 , the data to be shared (eg, first data) is relative to Figure 5 is called "specific input" to make it easier to compare with Figure 5 Distinguish from other data mentioned in the description.

[0073] At 591, method 590 includes receiving a request at a processing resource to share a specific input with a specific audience. For example, the request can be received as an upload at a sharing tool and can include different file types (e.g., pdf, word processing program, etc.), images, paragraphs from a specific file format, and / or circuit board files (e.g., circuitry, connections in memory types (e.g., NAND), etc.) and information (e.g., why the request is being made, supporting documentation, etc.). In some examples, the sharing tool and associated AI can recognize industry standard file types.

[0074] In some examples, the specific input may be categorized as text or image or both. Based on the categorization, the specific input may be analyzed using text recognition tools, image recognition tools, image / text separation tools, etc. to determine the content of the specific input.

[0075] At 592, method 590 includes comparing the specific input to a memory resource or storage device (e.g., a database) communicatively coupled to the processing resource to determine whether the specific input has been previously shared with the specific audience. At 593, method 590 includes determining a sharing status of the specific input based on the comparison, the sharing status including determining whether the specific input can be shared with the specific audience, cannot be shared with the specific audience, or whether it is undetermined whether the specific input can be shared with the specific audience.

[0076] For example, using AI associated with a sharing tool, if the comparison yields a match between the specific input and data previously deemed shareable with a specific audience, a "shareable" sharing state may be determined. If the comparison reveals a match between the specific input and data previously deemed unshareable with a specific audience, a "unshareable" sharing state may be generated. If neither of the aforementioned matches occurs, if a partial match occurs (e.g., same data with a different specific audience, some overlapping data, etc.), or if no match occurs, an "undetermined" sharing state may be generated.

[0077] At 594, method 590 includes notifying the sender of the request that the specific input can be shared with the specific audience in response to the sharing status indicating that the specific input can be shared. In this example, the specific input can be watermarked, and the watermarked specific input can be shared with the specific audience. At 595, method 590 includes notifying the sender of the request that the specific input cannot be shared with the specific audience in response to the sharing status indicating that the specific input cannot be shared.

[0078] At 596, method 590 includes requesting approval from an administrator to share the specific input with the specific audience in response to the undetermined sharing status, and notifying the sender of the result of the approval request. For example, one or more administrators (e.g., a group of administrators) may be prompted to determine the sharing status based on their knowledge of the subject matter, the specific input, the specific audience, and historical methods. The one or more administrators may determine that the specific input may be shared with the specific audience, may not be shared with the specific audience, or may determine that a portion of the specific input may be shared with the specific audience (e.g., portions must be deleted before sharing).

[0079] In some examples, a particular input may be classified as one of a plurality of predetermined security levels (e.g., "Highly Confidential," "Confidential," "Internal Only," "Public," etc.) based on the content of the particular input, and / or a confidence level may be assigned (e.g., using AI) to the security level classification. For example, if, based on training data, a particular input matches data in a pool previously classified as "Public" with a lower confidence level, the confidence level in the security level classification of the particular input may be lower than if the particular input matches data in a pool previously classified as "Public" with a higher confidence level. The first portion, the second portion, and the third portion may be labeled with an identifying mark based on their associated security level category, their associated confidence level, or both.

[0080] In some examples, the memory resource or storage device can be updated with the sharing status of a particular input in response to the sharing status indicating that the particular input can be shared or not shared, or with the approval / rejection decision from the administrator in response to the sharing status being undetermined. Information including who sent the request, the administrator who reviewed the request, and the reason for the decision can also be included in the update. These updates can be used in future data sharing requests because the sharing tool can use this information to self-learn.

[0081] Figure 6 is another flow chart illustrating an example method 687 for sharing data with a specific audience according to several embodiments of the present disclosure. The method 687 may be performed by, for example, Figure 3 and 4 The described system and other systems are implemented.

[0082] At 689, method 687 may include receiving first data at a processing resource. The first data may include specific input that is desired to be shared with a specific audience. For example, an engineer at a first company may desire to share a specific mechanical plan with an engineer at a second company.

[0083] At 691, method 687 may include determining, at a processing resource, whether the first data includes a combination of bits associated with text or an image, or both, and at 693, method 687 may include comparing the combination of bits with second data stored on a memory resource or storage device coupled to the processing resource via a network. At 695, method 687 may include identifying one or more words represented by the first data, one or more images represented by the first data, or both, based at least in part on the comparison. For example, the first data may be analyzed by comparison with the second data using a text recognition tool, an image recognition tool, an image / text separation tool, or the like to determine the content of the first data.

[0084] For example, if the first data is determined to be text, specific words in the text can be detected and / or identified, for example, using a text recognition tool. The words can be organized and a determination of their content can be made. If the first data is considered to be an image, one or more images can be reviewed and / or adjusted. For example, the images can be placed in the same orientation so that they are not tilted, rotated, etc. in different ways from each other. Each image can be segmented, and each segment can be analyzed to determine the content. If the first data is considered to be a combination of text and one or more images, an area (e.g., a rectangular area) that may be text can be detected within the first data, and the text can be separated from the image. Specific words in the text can be detected and / or identified, and one or more images can be reviewed and / or adjusted and segmented.

[0085] At 697, method 687 may include assigning first metadata representing a first security classification and a first confidence level based at least in part on the recognized text or words, and second metadata representing a second security classification and a second confidence level based at least in part on the recognized images or images to the first data. For example, metadata representing a security classification and a confidence level may be assigned to specific text, image segments, and images, as previously described herein. In some instances, third metadata representing one of a plurality of predetermined security levels based at least in part on the recognized text or words, the recognized images or images, or both may be assigned to the first data. Such predetermined security levels may include, for example, "Highly Confidential," "Confidential," "Internal Only," "Public," and the like. In some instances, fourth metadata representing one of the plurality of predetermined security levels based at least in part on the recognized text or words, the recognized images or images, or both may be assigned to one of the plurality of predetermined security levels. In some instances, the security level classification and / or confidence level may be labeled for ease of review.

[0086] In some examples, method 687 may include determining a sharing state of the first data (e.g., a "shareable" sharing state, a "non-shareable" sharing state, an "undetermined" sharing state, etc.) based at least in part on a comparison of a combination of bits with the second data, a first confidence level, and a second confidence level. The first data may be compared to a memory resource or storage device (e.g., a database) to determine whether an input having the same components has been previously shared with a particular audience. For example, a match is made if a particular word from the first data matches a word in the memory resource or storage device that has been previously shared with a particular audience. In some examples, only some particular words may match the memory resource or storage device, which may result in the deletion of the particular words that do not match. For example, if the particular word does not match a word in the memory resource or storage device that has been previously shared with a particular audience, or if the particular word does match a word in the memory resource or storage device but those words have not been previously shared with the particular audience, a match is not made. In some examples, the particular word may be matched to words in the memory resource or storage device that were previously not allowed to be shared with the particular audience. Based on these comparisons, the sharing state may be determined.

[0087] In some examples, method 687 may include updating a memory resource or storage device with the sharing status of the first data in response to the sharing status indicating that the first data is shareable or not shareable, and updating a memory resource or storage device of a processing resource coupled to the network with an approval decision from an administrator in response to the sharing status being undetermined. This may retrain the data used to make the sharing determination. For example, AI may be used to make the sharing status determination, and the updated memory resource or storage device may improve accuracy and efficiency.

[0088] In some examples, method 687 may include, in response to the sharing status indicating that the first data is shareable, notifying the sender of the first data that the first data is shareable with a specific audience; in response to the sharing status indicating that the first data is not shareable, notifying the sender of the first data that the first data is not shareable with the specific audience; and, in response to the sharing status being undetermined, requesting approval from an administrator to share the first data with the specific audience and notifying the sender of the result of the approval request. If the sharing status indicates that the first data is shareable, the first data may be watermarked and shared with the specific audience.

[0089] At 699, method 687 may include transmitting, via a network and at least in part based on the first or second security classification and the first or second confidence level, an output comprising the first data or third data to a set of users, the third data comprising a combination of bits modified relative to a combination of bits of the first data. For example, the third data may include the first data with deleted material or other modifications, or may include the first data with instructions to an administrator to review the first data and other data associated therewith (security classification, confidence level, etc.). In such examples, the set of users may include a specific audience, the sender of the first data, and / or an administrator or team of administrators designated to review the first data.

[0090] Although specific embodiments have been illustrated and described herein, it will be understood by those skilled in the art that arrangements calculated to achieve the same results may be substituted for the specific embodiments shown. The present disclosure is intended to cover adaptations or variations of one or more embodiments of the present disclosure. It will be understood that the above description is provided in an illustrative and non-restrictive manner. Combinations of the above embodiments and other embodiments not specifically described herein will be apparent to those skilled in the art upon reviewing the above description. The scope of one or more embodiments of the present disclosure includes other applications in which the above structures and processes are used. Therefore, the scope of one or more embodiments of the present disclosure should be determined with reference to the appended claims and the full scope of equivalents to which such claims are given.

[0091] In the foregoing Detailed Description, some features have been grouped together in a single embodiment for the purpose of streamlining the disclosure. This approach to the disclosure should not be interpreted as reflecting an intention that the disclosed embodiments of the disclosure necessarily utilize more features than are expressly recited in each claim. In fact, as reflected in the appended claims, the inventive subject matter lies in less than all features of a single disclosed embodiment. Therefore, the appended claims are hereby incorporated into the Detailed Description, with each claim standing on its own as a separate embodiment.

Claims

1. A method (687) for sharing data, comprising: receiving, at a processing resource (362, 476), first data (689); determining, at the processing resource (362, 476), whether the first data includes a combination of bits (691) associated with text or an image or both; comparing the combination of bits to second data (693) stored on a memory resource (360, 474) or storage device coupled to the processing resource (362, 476) via a network; identifying one or more words represented by the first data or one or more images represented by the first data, or both, based at least in part on the comparison (695); assigning to the first data first metadata representing a first security classification and a first confidence level based at least in part on the identified one or more words, and second metadata representing a second security classification and a second confidence level based at least in part on the identified one or more images (697); transmitting, via the network and based at least in part on the first or second security classification and the first or second confidence level, an output comprising the first data or third data to a set of users, the third data comprising a combination of bits modified relative to the combination of bits of the first data (699); determining a sharing status of the first data based at least in part on the comparison of the combination of bits with the second data, the first confidence level, and the second confidence level; updating the memory resource (360, 474) or storage device coupled to the processing resource (362, 476) via the network with the sharing status of the first data in response to the sharing status indicating that the first data is shareable or not shareable; as well as The memory resource (360, 474) or storage device coupled to the processing resource (362, 476) via the network is updated with an approval decision from an administrator in response to an undetermined sharing status.

2. The method (687) of claim 1, further comprising assigning to the first data third metadata representing one of a plurality of predetermined security levels based at least in part on the identified one or more words, the identified one or more images, or both.

3. The method (687) of claim 2, further comprising assigning to said one of said plurality of predetermined security levels fourth metadata based at least in part on said identified one or more words, said identified one or more images, or both.

4. The method (687) of claim 1, further comprising: in response to the sharing status indicating that the first data is shareable, notifying a sender of the first data that the first data can be shared with a specific audience; in response to the sharing status indicating that the first data is not shareable, notifying the sender of the first data that the first data is not shareable with the specific audience; as well as In response to the undetermined sharing state, the administrator is requested to approve the sharing of the first data with the specific audience, and the sender is notified of the result of the approval request.

5. The method (687) of claim 1, wherein in response to the sharing status indicating that the first data is shareable: adding a watermark to the first data; and The watermarked first data is shared with a specific audience.

6. A non-transitory machine-readable medium comprising a processing resource (362, 476) in communication with a memory resource (360, 474) having instructions executable to: receiving a request (478) to share first data received at the processing resource (362, 476) with a specific audience; determining, using artificial intelligence (AI), based on a comparison of the request with second data stored on the memory resource (360, 474) communicatively coupled to the medium, a first portion of the first data that is shareable with the specific audience, a second portion of the first data that is not shareable with the specific audience, a third portion of the first data that has an undetermined sharing status, or a combination thereof (480); assigning a security level category to the first portion, the second portion, and the third portion; assigning a confidence level to the security level classification using the AI; requesting approval from an administrator to share the third portion with the specific audience (482); writing data associated with the request, the first portion, the second portion, and the third portion to the memory resource (360, 474, 484); watermarking the first and third portions and sharing the first and third portions with the specific audience in response to the third portion being approved for sharing (484); and The first portion is watermarked and shared with the specific audience in response to the third portion being approved for sharing (486).

7. The non-transitory machine-readable medium of claim 6, further comprising instructions executable to determine whether the first data includes a combination of bits associated with text or an image or both.

8. The non-transitory machine-readable medium of claim 6, wherein the first data associated with the request, the first portion, the second portion, and the third portion in the memory resource (360, 474) includes whether the first data has been shared, whether the first data is currently being shared, whether the first data is planned to be shared, with whom the first data has been or is planned to be shared, who requested sharing of the first data, the administrator, and keywords associated with the first data, or a combination thereof.

9. The non-transitory machine-readable medium of any one of claims 6 to 8, further comprising instructions executable to update the memory resource (360, 474) using written data in the memory resource (360, 474) associated with the request, the first portion, the second portion, and the third portion.

10. The non-transitory machine-readable medium of any one of claims 6 to 8, further comprising instructions executable to: receiving input regarding the performance of the medium from the party sending the request, the specific audience, or a combination thereof; and The performance input is implemented into the medium.

11. The non-transitory machine-readable medium of claim 6, further comprising instructions executable to perform the following operations: labeling the first part, the second part, and the third part with an identification mark based on their associated security level categories, associated confidence levels, or both.

12. The non-transitory machine-readable medium of any one of claims 6 to 8, further comprising instructions executable to use the AI ​​to recognize an industry standard file type associated with the first data.

13. The non-transitory machine-readable medium of claim 6, wherein the instructions executable to determine the first portion, the second portion, and the third portion comprise instructions executable to: categorizing the first data using the AI, wherein the first data has a combination of bits associated with text or images or both (364); using the AI ​​to classify the first data into one of a plurality of predetermined security levels based on the analyzed content of the classified first data (366); assigning a confidence level to the security level classification using the AI ​​(368); comparing (370) the combination of bits with second data stored on the memory resource (360, 474); as well as A sharing status of the first portion, the second portion, and the third portion of the first data is determined based on the comparison, the security level classification, and the confidence level (372).

14. The non-transitory machine-readable medium of claim 13, further comprising instructions executable to: In response to the sharing status indicating that the first portion of the first data is shareable, notifying the sender of the request that the first portion of the first data is shareable with the specific audience; and In response to the sharing status indicating that the second portion of the first data is not shareable, the sender of the request is notified that the second portion of the first data is not shareable with the specific audience.

15. The non-transitory machine-readable medium of any one of claims 13 to 14, wherein in response to the classification as text, the instructions are further executable to: Detecting specific words or word groups in the text; using the AI ​​to classify the specific text or text group into a corresponding one of the plurality of security levels; and The AI ​​is used to assign a confidence level to each of the assigned security levels.

16. The non-transitory machine-readable medium of any one of claims 13 to 14, wherein in response to the classification as an image, the instructions are further executable to: adjusting the orientation of the image; segmenting the image into a plurality of segments; classifying each of the plurality of segments into a corresponding one of the plurality of security levels using the AI; and The AI ​​is used to assign a confidence level to each of the assigned security levels.

17. The non-transitory machine-readable medium of any one of claims 13 to 14, wherein in response to the classification as a combination of text and an image, the instructions are further executable to: separating the first data into text and images; Detecting specific words or word groups in the text; adjusting the orientation of the image; segmenting the image into a plurality of segments; using the AI ​​to classify the specific word or word group and each of the plurality of segments into a corresponding one of the plurality of security levels; and The AI ​​is used to assign a confidence level to each of the assigned security levels.

Images