US20210357439A1 - Gathering information on user interactions with natural language processor (nlp) items to determine an order in which to present nlp items - Google Patents

Abstract

Provided are a computer program product, system, and method for providing information on detected user interactions with natural language processor (NLP) items to a server to use to determine an order in which to render NLP items in a user interface. Interactions with natural language processing (NLP) items in documents are detected. For each interaction with an NLP item of the NLP items in the documents, determination is made of a context attribute value for a context attribute related to the interaction with the NLP item. Then interaction information is generated indicating the NLP item and the determined context attribute value. The interaction information is sent to the server to include in user interaction information in a database.

Description

BACKGROUND OF THEINVENTION1. Field of the Invention
• The present invention relates to a computer program product, system, and method for gathering information on user interactions with natural language processor (NLP) items to order presentation of NLP items in documents.
2. Description of the Related Art
• Natural language processing (NLP) of unstructured text documents allows a software system to extract concepts from the documents that can then be analyzed in order to present meaningful information to a user. For instance, medical documents (physician clinic notes, pathology reports, etc.) may be processed by the NLP system in order to extract a patient's diagnosis, past medical history, therapy history, and other data needed to evaluate a patient for the next steps in their treatment. With NLP processing, there may be errors in the processed data that may need to be corrected. Mistakes may be especially problematic if the errors are for mission critical information, like errors in medical information, such as patient medical conditions, diagnosis, current medication regimens, etc.
• There is a need in the art to provide improved techniques for presenting users with NLP extracted information to review to optimize user interaction to review and correct errors in the NLP content.
SUMMARY
• Provided are a computer program product, system, and method for providing information on detected user interactions with natural language processor (NLP) items to a server to use to determine an order in which to render NLP items in a user interface. Interactions with natural language processing (NLP) items in documents are detected. For each interaction with an NLP item of the NLP items in the documents, determination is made of a context attribute value for a context attribute related to the interaction with the NLP item. Then interaction information is generated indicating the NLP item and the determined context attribute value. The interaction information is sent to the server to include in user interaction information in a database.
BRIEF DESCRIPTION OF THE DRAWINGS
• FIG. 1 illustrates an embodiment of a network information management environment.
• FIG. 2 illustrates an embodiment of an interaction information package sent form a user computing device on interactions with Natural Language Processor (NLP) items.
• FIG. 3 illustrates an embodiment of user interaction information on interaction with NLP items.
• FIG. 4 illustrates an embodiment of NLP item priority information.
• FIG. 5 illustrates an embodiment of an NLP item transmission.
• FIG. 6 illustrates an embodiment of operations at an interacting user computing device to detect interactions with NLP items and generate the interaction information package.
• FIG. 7 illustrates an embodiment of operations to process an interaction information package to generate user interaction information.
• FIG. 8 illustrates an embodiment of operations to process user interaction information in a user interaction database to generate priorities for NLP items used to determine an order for presenting NLP items to a reviewing user.
• FIG. 9 illustrates an embodiment of operations to detect and send information on user interaction with NLP items and to render NLP items in a document in a user interface according to an ordering based on the user interactions.
• FIG. 10 illustrates an embodiment of operations to process modifications to NLP items from the reviewing user.
• FIG. 11 illustrates a computing environment in which the components ofFIG. 1 may be implemented.
DETAILED DESCRIPTION
• NLP systems may need to present to user NLP items and content in a document to review or correct. However, users in a system may have limited to review NLP items in a document to correct. In a medical record system, a doctor may be presented with a vast array of important NLP items in medical records to correct. This can consist of many screens of information, and it is often overwhelming to a user. In fact, a user may become so frustrated with having to correct so many NLP items, they may decide to avoid using the NLP correction and review system due to the time commitment.
• Described embodiments provide techniques to observe user interaction with NLP items in a document to determine the interaction frequency with the NLP items and context attribute values related to interaction with the document. The gathered NLP item interaction information, along with context attribute values related to the context of the interaction, may be saved as user interaction in a computer database. When the NLP system is generating a document to present to a user, the NLP system may then consider interaction frequencies of NLP items in specific contexts, similar to the context with which the document will be reviewed, to determine a priority of NLP items based on their interaction frequency. This priority information may be used to determine an order in which the NLP items are presented to a user to review and correct, so that the user will focus on those NLP items collectively deemed by similar users in similar contexts in the system to be of high importance, as evidenced by their interaction frequency with the NLP items. The user may then focus on those NLP items they recognize as important and ignore or disregard the later presented NLP items having lower priority, as evidenced by a lower interaction frequency.
• Described embodiments also provide improved techniques and data structures for gathering NLP item interaction information at the user computing devices to record the relevant information concerning the user interaction with the NLP items rendered in a computer user interface, including context attribute values related to the context of the interaction and the interaction frequency with the NLP items.
• Described embodiments also provide improved computer database structures to improve the management of information on user interactions with NLP items and the context of that information to allow later database searching of the interaction information to generate priority values for the NLP items that are specific to the context attributes in which the interactions occurred. Examples of context attributes include user specific context, subject matter specific context of the NLP items, and NLP item specific attributes, such as interrogative sentences in which the NLP items were included. These improved database structures allow for focus on accessing NLP item information for specified context attribute values to use to determine a priority based on the interaction by users across documents in the network in similar context. The cumulative priority may then be used to determine an order in which NLP items should be presented to the user to review that reflects priority based on actual interaction by similarly situated users with the NLP items.
• Described embodiments provide further improved computer technology to improve the way the user interface program of the user computer gathers user interactions with NLP items to provide to the server and to present NLP items to the user so as to optimize the user responses and corrections to NLP items in a document. With described embodiments, user computers detect interactions with natural language processing (NLP) items in documents and, for each interaction with an NLP item, each user computer sends interaction information to the server on the interaction including information on a context attribute value for a context attribute related to the interaction with the NLP item.
• With described embodiments, multiple of the user computers gather their information on interactions with NLP items to send to the server to use to determine a priority value for the NLP items. When a user requests a document from the server, the server will provide an order for the NLP items in the requested document based on the NLP item interaction information from the plurality of computers. The user interface at the user computers then presents the NLP items in a requested document to the user in the order, determined according to the collective information on user interactions with the NLP items gathered by the user computers.
• By presenting NLP items having errors to correct to a user in the user interface according to an order based on a priority determined by a frequency of access by similarly situated users to the NLP items priority of the NLP items, the system increases the likelihood the user responds to the highest priority NLP items to correct by first presenting those higher priority NLP items early in the ordering. In this way, the user is not overwhelmed by being presented with all the NLP items to correct, but is provided NLP items in an order determined by a frequency of access by similarly situated users to the NLP items. The assists the user in focusing on first correcting those NLP items in the document having a highest priority, and when the user no longer has time or interest in correcting NLP items, those lower priority NLP items will be ignored, while the highest priority NLP items presented first for consideration to the user will likely have been corrected.
• FIG. 1 illustrates an embodiment of an information retrieval environment having aninformation manager server100 maintaining auser interaction database102 having information on user interaction withdocuments104 in thedatabase102. Theinformation manager server100 may communicate with a plurality ofuser computing devices106₁,106_i. . .106_nover anetwork108, where the users are recognized by theinformation manager server100 to interact with thedocuments104. Theinformation manager server100 includes amemory110 including anoperating system112 to manageserver100 operations and interact with connected devices; anatural language processor114, such as by way of example, the Watson™ Natural Language, to process thedocuments104 to determine Natural Language Processor (NLP) items in thedocuments104, which may comprise entities identified in thedocuments104, such as classified named entities in the text classified into categories, such as names of persons, organizations, locations, expressions of times, terms and entities related to the subject matter of thedocuments104, etc.; adocument manager116 to manage access todocuments104 in theuser interaction database102; and aninteraction analyzer118 to analyze and process user interaction information included in aninteraction information package200 transmitted from theuser computing devices106₁,106_i. . .106_n.
• Theuser interaction database102 may include records and data structures, shown as maintained in thememory110, includinguser profile information120 having information on the users of theuser computing devices106₁,106_i. . .106_n, including profiles related to positions or roles of the users in the environment in which theinformation manager server100 is deployed;subject information122 having information on subjects of thedocuments104, i.e., to which the documents pertain104, e.g., the subjects may comprise patients and thedocuments104 may comprise patient medical records;user interaction information300 generated by theinteraction analyzer118 to maintain information on user interaction with NLP items indocuments104; and NLPitem priority information400 having priority values for the NLP items calculated by theinteraction analyzer118 and used to determine an order in which the NLP items are presented to the users at theuser computing devices106₁,106_i. . .106_nto consider. Theinteraction analyzer118 may generate anNLP item transmission500 for a document providing an order with which to present the NLP items in the document for the user to review.
• Theuser interaction database102 may comprise a relational or object oriented database, or other suitable database types.
• In one embodiment, thedocuments104 may comprise medical records and the subjects may comprise patients for which themedical records106 are maintained. Amedical record106 may include numerous interrogatory sentences asking questions about the subject/patient's medical history, personal history, current diagnosis, current medication, etc., comprising the NLP items. Thus the NLP items may comprise objects of the interrogatory sentences or other patient and medical information the users enter into the medical records. The users may comprise doctors, nurses, and other medical professionals that may add information to the medical records. Thedocuments104 may comprise a structured or unstructured document, and be in a format such as Extended Markup Language (XML), Hypertext Markup Language (HTML), a text format, word processor format, etc.
• Theuser computing devices106₁,106_i. . .106_nmay include, as shown withuser computing device106_i, anoperating system124 to manage and interact with (Input/Output I/O) devices126, auser interface128 in which to renderdocuments104 from theinformation manager server100 and allow the user to interact with thedocuments104 with the I/O devices126; andinteraction detection unit130 to detect user interactions with the NLP items in thedocument104 rendered in theuser interface128 with the I/O devices132; andinteraction information package200 generated by theinteraction detection unit130, including information on user interaction with NLP items in adocument104 rendered in theuser interface128.
• In one embodiment the I/O devices126 may comprise a mouse, keyboard, etc., and the interactions may comprise the user selecting or entering content for the NLP items in thedocument104 rendered in theuser interface128. The I/O device126 may further comprise a gaze tracking device to determine an NLP item in the document or interrogative sentence including the NLP item at which the user is gazing. A gaze tracking device is typically worn as glasses and includes cameras on the glasses to acquire gazed text or images being gazed by the eyes of the user. The gaze tracking device includes a camera to capture and analyze an eye image to determine the pupil location to acquire the gazed image or text the user eyes are staring at or tracking, i.e., gazing at directly. Theinteraction detecting unit130 may determine the NLP item or interrogative sentence including the NLP item with which the user is interacting, such as selecting with a mouse pointer, entering text for with a keyboard, or gazing at with a gaze tracking glasses.
• Theuser computing devices106₁,106_i. . .106_nmay comprise a smart phone, tablet, personal digital assistance (PDA), laptop, or stationary computing device, e.g., desktop computer, server. Thememory110 may comprise non-volatile and/or volatile memory types, such as a Flash Memory (NAND dies of flash memory cells), a non-volatile dual in-line memory module (NVDIMM), DIMM, Static Random Access Memory (SRAM), ferroelectric random-access memory (FeTRAM), Random Access Memory (RAM) drive, Dynamic RAM (DRAM), storage-class memory (SCM), Phase Change Memory (PCM), resistive random access memory (RRAM), spin transfer torque memory (STM-RAM), conductive bridging RAM (CBRAM), nanowire-based non-volatile memory, magnetoresistive random-access memory (MRAM), and other electrically erasable programmable read only memory (EEPROM) type devices, hard disk drives, removable memory/storage devices, etc.
• The userinteractive database102 may be implemented in one or more storage devices, such as magnetic hard disk drives, solid state storage device (SSD) comprised of solid state electronics, EEPROM (Electrically Erasable Programmable Read-Only Memory), flash memory, flash disk, Random Access Memory (RAM) drive, storage-class memory (SCM), etc., Phase Change Memory (PCM), resistive random access memory (RRAM), spin transfer torque memory (STT-RAM), conductive bridging RAM (CBRAM), magnetic hard disk drive, optical disk, tape, etc. Data in thestorages102₁,102₂. . .102_n4 may further be configured from an array of devices, such as Just a Bunch of Disks (JBOD), Direct Access Storage Device (DASD), Redundant Array of Independent Disks (RAID) array, virtualization device, etc.
• Thenetwork108 may comprise one or more networks including Local Area Networks (LAN), Storage Area Networks (SAN), Wide Area Network (WAN), peer-to-peer network, wireless network, the Internet, etc.
• Generally, program modules, such as theprogram components112,114,116,118,124,128,130, and any others described herein, may comprise routines, programs, objects, components, logic, data structures, and so on that perform particular tasks or implement particular abstract data types. The program modules may be practiced in distributed cloud computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed cloud computing environment, program modules may be located in both local and remote computer system storage media including memory storage devices.
• The program components and hardware devices of theuser computing devices106₁,106_i. . .106_nandinformation management server100 ofFIG. 1 may be implemented in one or more computer systems, where if they are implemented in multiple computer systems, then the computer systems may communicate over a network.
• Theprogram components112,114,116,118,124,128,130, and any others described herein, may be accessed by a processor from memory to execute. Alternatively, some or all of theprogram components112,114,116,118,124,128,130, and any others described herein, may be implemented in separate hardware devices, such as Application Specific Integrated Circuit (ASIC) hardware devices.
• The functions described as performed by theprogram components112,114,116,118,124,128,130, and any others described herein, may be implemented as program code in fewer program modules than shown or implemented as program code throughout a greater number of program modules than shown.
• FIG. 2 illustrates an embodiment of aninteraction information package200_igenerated by theinteraction detection unit130 in auser computing device106_iand includes auser identifier202 of the user interacting with theuser interface128 when theinteraction information package200_iwas generated; anNLP item204 for which the interaction was detected; one or more context attribute values206 for context attributes related to the user interaction with theNLP item204 and aninteraction frequency208 indicating one or more distinct interactions with theNLP item204.
• FIG. 3 illustrates an embodiment ofuser interaction information300_igenerated by theinteraction analyzer118 from a receivedinteraction information package200_i, and includes auser identifier302 of the user interacting with theNLP item306, which may be optional; andinteraction ID304 uniquely identifying theuser interaction information300_i; theNLP item306 subject to the interaction; one or more profile attribute values308 of profiles of theuser302, such as role in an organization, position, personal identification information, etc.; one or more context attribute values310 from those206 included in theinteraction information package200; and aninteraction frequency312 also from thecorresponding field208 in theinteraction information package200_i.
• FIG. 4 illustrates an embodiment of an instance of NLPitem priority information400_i, which may be generated at the time adocument104 is selected to provide to a user to review or may be generated in advance to be available when a document is selected, and periodically updated. The NLPitem priority information400_iincludes theNLP item402 for which the value is generated; aprofile attribute value404 of the users whose interactions were considered to calculate the priority value; one or more context attribute values406 related to the context in which users whose interactions were considered to calculate the priority value interacted with theNLP item402; acumulative interaction frequency408 comprising the total or combination of theinteraction frequencies312 from allinteraction information instances300_ihavingNLP item306,profile attribute value308, and context attribute values310 matching those402,404, and406 of thepriority value410 being calculated; and apriority value410 derived from thecumulative interaction frequency408. Thepriority value410 may be determined based on a priority range the number ofinteractions408 falls within, high, medium, low, etc., or may be calculated as a percentage of the interactions for anNLP item402 across context attributes and profiles, or within context attributes and profiles.
• FIG. 5 illustrates an embodiment of an instance of anNLP item transmission500_isent to auser computing device106₁,106_i. . .106_n, including indication or a copy of a selecteddocument502 for which the transmission applies and for each of the NLP items within the selecteddocument502, a NLP item/content pair504₁,504₂. . .504_nand an order value506₁,506₂. . .506_nindicating an order in which the NLP item will be presented to the user in theuser interface128. TheNLP item504_iinformation may further indicate a location in the selecteddocument502 of theNLP item504_i.
• FIG. 6 illustrates an embodiment of operations performed by theinteraction detection unit130 in auser computing device106_ito provide theinformation management server100 with information on a user interaction at theuser computing device106_iwith an NLP element in adocument104 rendered in theuser interface128. Upon theinteraction detection unit130 detecting (at block600) an interaction with an NLP item rendered in theuser interface128 through an I/O device126, theinteraction detection unit130 determines (at block602) context attribute values for context attributes related to the interaction, including user attributes of the user interacting with the NLP item (e.g., location, time, cohorts or groups in which the user is included), a question subject attributes having information on a subject of an interrogative sentence including the NLP item (e.g., age, profession, etc.), and NLP specific attribute values, such as an interrogative sentence including the NLP item. For instance, the location may be determined from a location indicated in theuser computing device106_i, such as from stored information or information from a Global Positioning System (GPS) device, and the time may be determined from the system time at thedevice106_i. Context attribute values based on the content of thedocument104 being rendered, such as an interrogative sentence, may be determined by using NLP algorithms to analyze the sentence structure including the NLP item subject to the interaction. The interaction frequency with the NLP item is determined (at block604), which may be one or more times if there are multiple distinct interactions as determined by theinteraction detection unit130.
• A user interaction may comprise the user entering information for an NLP item, such as an answer to an interrogative question, considering the NLP item, as evidenced by detection of mouse/keyboard interaction or gazing at a sentence including the NLP item.
• Theinteraction detection unit130 generates (at block606) aninteraction information package200_iincluding auser ID202 of the user of theuser computing device106_icomputing unit, theNLP item204 subject to the interaction and the determined context attribute values206 andinteraction frequency208. Theinteraction information package200_iis then transmitted (at block608) to theinformation manager server100. Aninteraction information package200_imay include information for just interaction with one or multiple NLP items in thedocument104.
• With the embodiment ofFIG. 6, theuser computing device106_igathers directly information on how the user interacts with the NLP items, which may comprise answers to an interrogative sentence. This information may then be combined with such interaction information from multiple users to determine an importance of an NLP item as evidenced by cumulative user interaction with that item. Theinformation manager server100 may then use this interaction information to determine priorities for NLP items to determine an order in which to present NLP items to the users to review at a later time, so that users are provided first NLP items having a highest interaction rate or considered most by users to optimize the user response rate to review NLP items. This provides users with NLP items to review based on their own previous selection of NLP items they prioritized when they were reviewing other documents. Presenting to a user NLP items to consider in an order based on the rate of interaction with NLP items by the user and their cohorts increases the likelihood that the user will review the presented NLP items, because they are presented in an order of importance determined by their own previous considerations and interactions.
• FIG. 7 illustrates an embodiment of operations performed by theinteraction analyzer118 to process a receivedinteraction information package200_i. Upon receiving (at block700) aninteraction information package200_ifrom an interacting user at acomputing device106₁, theinteraction analyzer118 generates (at block702) auser interaction information300_iinstance having the user ID302 (optional) of the interacting user at theuser computing device106_ithat sent thepackage200_i, aninformation instance ID304, a profile attribute value of the user sending thepackage200_i, which may be determined from theuser profile information120 or sent in thepackage200_i, and theinteraction frequency312 as indicated infield208 of thepackage200_i. Eachcontext attribute value206 in theinteraction information package200_iis indicated (at block704) in acontext attribute value310 filed in the userinteraction information instance300_i. The generateduser interaction information300_imay then be stored (at block706) in theuser interaction database102, such as a database record or object.
• With the embodiment ofFIG. 7, theinformation manager server100 maintains information on interactions with each NLP item in different database records across all users in the system to allow for database operations to search and combine information based on NLP items, user profiles, and context attribute values.
• FIG. 8 illustrates an embodiment of operations performed by theinteraction analyzer118 to determine an order in which to present NLP items in a selecteddocument104 to be viewed in theuser interface128 of auser computing device106_iof a reviewing user. This operation may be initiated when a reviewing user selects adocument104 to review. Upon processing (at block800) a selected document having NLP items, theinteraction analyzer118 calls (at block802) theNLP114 to determine NLP items in the selecteddocument104, such as any NLP item or NLP items having content to review or with errors to correct. A profile value of the reviewing user to which thedocument104 will be sent is determined (at block804), which may be determined from theuser profile information120, such as a role of the reviewing user in the organization. A subject of the selecteddocument104 is determined (at block806), this may be a person or entity associated with thedocument104, such as a patient associated with a medical record of their medical information. The subject may also be determined by processing the document to determine if a person is identified as the subject of the document, such as the subject of an interrogative question requesting the identity of the person for which thedocument104 is provided.
• Theinteraction analyzer118 determines (at block808) current context attribute values for context attributes, including user context attributes, such as a current time, the current location of theuser computing device106_i, cohort values of groups of users in which the user is included, such as an experience level, specific job title, etc., and subject attribute values for the subject of the document, such as personal information on the person that is the subject matter of thedocument104, e.g., age, personal or sensitive information of relevance, etc.
• Theinteraction analyzer118 then performs a loop of operations atblock810 through820 for each NLP item in the selecteddocument104. Atblock812, theinteraction analyzer118 determines NLP item specific attributes in the selecteddocument104, such as an interrogative sentence or other information specific to the NLP item being considered. Theinteraction analyzer118 may then initiate a query of theuser interaction database102 to determine (at block814) userinteraction information instances300_ihaving theprofile attribute value308 and current context attribute values310 for theNLP item306 being considered, e.g., user attribute values, subject attribute value, and NLP specific attribute values, etc., to access the userinteraction information instances300_irelevant to the current user access of thedocument104. Theinteraction analyzer118 determines (at block816) acumulative interaction frequency408 for the NLP item based on theinteraction frequency312 for each of the determined relevantinteraction information instances300_i. Thecumulative interaction frequency408 may comprise the sum ofinteraction frequencies312 for all the determined interaction information instances, or comprise some derived value, such as the some of theinteraction frequencies312 weighted or as a percentage of interactive frequencies across profile or context attributes for the NLP item.
• Theinteraction analyzer118 may then assign (at block818) apriority value410 for the NLP item based on thecumulative interaction frequency408, and generate NLPitem priority information400_i, having theNLP item402 and fields/parameters used to search, includingprofile attribute value404 and context attribute values406, the determinedcumulative interaction frequency408 andpriority value410. This NLPitem priority information400_imay be stored in theuser interaction database102 or maintained temporarily and continually recalculated when needed for adocument104 to provide to a reviewing user.
• After determining the priority values410 for all the NLP items in thedocument104, theinteraction analyzer118 determines (at block822) an order in which to present NLP items based on the priority values410 assigned to the determined NLP items. For instance, NLP items with higher priorities would be presented to the reviewing user before NLP items with a lower priority or lower level of interaction. AnNLP item transmission500_iis generated (at block824) for the selecteddocument502 having NLP items andcontent504₁,504₂. . .504_nalong with the determined order values506₁,506₂. . .506_nto control theuser interface128 to present NLP items in the determined order. The generatedNLP item transmission500_iis then transmitted (at block826) to theuser computing device106_i. Theuser interface128 would use the order values506₁,506₂. . .506_nto determine the order in which to present the NLP items andcontent504₁,504₂. . .504_nin the selecteddocument502 to the user to review. In one embodiment, theNLP item transmission500_imay not include NLP items of lower priority so as not to burden the reviewing user with too many items to consider. In this way, theNLP item transmission500_imay only include a subset of the NLP items in thedocument502 having ahigh priority value410. Alternatively, the user may be presented with the lower priority NLP items later at a time they are less likely to continue with their review.
• The described embodiments ofFIG. 8 provide a database driven computer solution to maintaining information on user interaction with NLP items in documents to use to determine an order in which to present to the NLP items in a selected document to review. Described embodiments determine priorities of NLP items based on interactions with those same NLP items across users. Described embodiments may further only consider those interactions having context attributes that are relevant to the current context attributes for the presentation of the document. In this way, interaction frequency of NLP items being considered to determine the NLP item priority is specific to the current attributes that are present during the current document viewing, such as user specific context attributes, location, time, cohorts, and attributes specific to the subject of thedocument104, such as patient or person to which the document pertains, as well as NLP specific items, such as an interrogative sentence in which the NLP item is considered. In this way the current calculated priority for the NLP items are specific to the use in which the document will be presented and will accurately reflect the priority that the user will expect.
• FIG. 9 illustrates an embodiment of operations performed by theuser interface128 of theuser computing device106_ito render NLP items in adocument502 provided with theNLP item transmission500_i. Upon receiving (at block900) from theinformation manager server100 anNLP item transmission500_ihaving adocument502 andNLP items504₁. . .504_nand an order506₁. . .506_nof the NLP items in thedocument502, theuser interface128 sets (at block902) a variable i to one. The ith orderedNLP item504_ihaving the order506₁is accessed in theNLP transmission500_i. Indication is made (at block906) in theuser interface128 of the ith ordered NLP item for the user to consider, such as by moving a cursor or other data entry graphical element to the ithorder NLP item504_iin theuser interface128 to enable the user to enter content for theNLP item504_i. Any user modification to the content for theNLP item504_iindicated in theuser interface128 is saved (at block908).
• Upon receiving (at block910) user selection to move to a next NLP item in thedocument502, the variable i is incremented (at block912) and if (at block914) there is a next (i+1)th NLP item504_i+1in the order506_i+1, then control proceeds to block906 to indicate that next NLP item. If all NLP items in the order of theNLP transmission500_ihave been considered, which may comprise just a subset of the NLP items in thedocument502, then theuser computing device106_imay send (at block916) the modified content for the NLP items for thedocument502 to theinformation management server100 to store for thedocument502/104.
• With the embodiment of operations forFIG. 9, theuser interface128 may guide the user to the NLP items in the document according to the order based on a priority assigned to the NLP items, which is calculated based on the frequency of access of the NLP items by users in a context and having similar roles to the context/user role at theuser computing device100_i. In this way, theuser interface128 is able to direct the user to those NLP items to correct having most priority or relevance to similarly situated users in the system to optimize the user response to correcting the NLP items.
• FIG. 10 illustrates an embodiment of operations performed by thedocument manager116 to process changes to content for NLP items received from auser computing device106_iin response to the user reviewing NLP items presented in an order according to the interaction priority. Upon receiving (at block1000) the user modifications to content for NLP items in the selecteddocument104 presented to the user, thedocument manager116 modifies (at block1002) the content for the NLP items in thedocument104 with the received modifications to the content for the NLP items. The modifieddocument104 is then saved (at block1004), such as in theuser interaction database102.
• With the described embodiments, NLP items with a lower priority may not be presented to the reviewing user to review. However, the user may also be able to access the full set of NLP items that are not prioritized as high. The observation and tracking of NLP item interaction can be performed at various user levels. At a single user level, the interactions of a single user may be observed. For instance, a doctor may frequently check and correct an NLP item with a certain attribute, but also have to access additional NLP items at the same time. All these interactions would be recorded. Further, interactions with NLP items may be considered at a group level, such interactions of a group of users sharing a group profile attribute. For instance, in a medical environment, the pulmonologists may focus on interacting with respiratory related medical information, whereas an oncologist may focus on cancer related NLP items. Other examples of groupings of users could be pathologists that verify and correct certain pathology items, and surgeons that verify and correct other surgery-related items. User roles can be determined using a staff directory such as in theuser profile information120.
• NLP items can also be considered that have certain additional attribute contexts, such that NLP items may only be considered that have a location attribute of the location of the reviewing user may be considered. For instance, when a doctor is in a clinical workroom meeting with patients, the doctor may only interact with a few NLP items in the patient record. However, the doctor in their office or at home may interact with NLP items in greater detail because in such settings they have more time to carefully consider patient information. Furthermore, the doctor may work at multiple clinic locations, one of which is the primary hospital treating patients with critical needs, and an alternate regional location that is used for regular follow-up with patients that are not in an urgent care situation, and the doctors consistent differences in interactions with NLP items at the different locations would be observed and recorded in user interaction information instances. This allows customization of the priority of NLP items based on their location.
• NLP item interaction frequency may also be distinguished based on the time of day at which the user is interacting with the system. For a doctor, the beginning of the day at the clinic, such as cancer center, can be hectic, and the doctor may have very limited time to review NLP items in the medical report as they have to meet with many patients. At other times of the day, such as in the afternoon, the doctors may have more time to work with a single patient, and may review more details and NLP items in the medical record. The described embodiments record the patterns of what attributes affect the interaction frequency of different NLP items for different times of a day.
• An age of patient context attribute could be considered when determining what NLP data is reviewed and corrected by the doctor. The system may record that when pediatric patients are involved, the doctor prefers to review and correct certain NLP items, versus for elderly patients. Described embodiments can observe the interactions and classify NLP item review and corrections needs based on observing what the doctor reviews and corrects for different ages of patients.
• A cohort of user attributes may provide a more fined grained categorization of a group in which the reviewing user is considered. For instance, in addition to recognizing that a doctor is a Leukemia doctor, there are more fine grained categorizations of cohorts of users, such as “leukemia doctors with 10 or more years of experience”, “leukemia doctors who specialize in Acute Myeloid Leukemia (AML) leukemia patients”, “general oncologist who sees leukemia, lung, melanoma, and many other cancer types”, other classifications of users may also be provided that are observed to have a correlation with interactions with NLP items. The system could observe the NLP corrections for each user, and over time may be able to see patterns that for a certain cohort of users (e.g. AML leukemia specialists), certain NLP correction and review is common
• Upon making these observations over time, the NLP items that are frequently corrected by a user, or by a particular user-role (such as Leukemia doctors), can be prioritized higher as items that should be presented to the user in that cohort of users. Likewise, items that are infrequently reviewed or corrected may be prioritized lower over time because they are generally considered less by this user or user-role. Theuser interface128 can monitor which NLP items are modified and based on the frequency count, can see which types of data are updated. The high priority NLP derived attributes can be presented to the user as the items that need immediate review. The user could also be given an option to see additional NLP derived attributes, but the user could focus on those high priority items if they have limited time. If the system has no high priority NLP derived attributes that need correction (that is, it has confidence that it already has the data it needs), then the user would not be prompted for any NLP corrections.
• The reference characters used herein, such as i and n are used to denote a variable number of instances of an element, which may represent the same or different values, and may represent the same or different value when used with different or the same elements in different described instances.
• The present invention may be a system, a method, and/or a computer program product. The computer program product may include a computer readable storage medium (or media) having computer readable program instructions thereon for causing a processor to carry out aspects of the present invention.
• The computer readable storage medium can be a tangible device that can retain and store instructions for use by an instruction execution device. The computer readable storage medium may be, for example, but is not limited to, an electronic storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. A non-exhaustive list of more specific examples of the computer readable storage medium includes the following: a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), a static random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanically encoded device such as punch-cards or raised structures in a groove having instructions recorded thereon, and any suitable combination of the foregoing. A computer readable storage medium, as used herein, is not to be construed as being transitory signals per se, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through a waveguide or other transmission media (e.g., light pulses passing through a fiber-optic cable), or electrical signals transmitted through a wire.
• Computer readable program instructions described herein can be downloaded to respective computing/processing devices from a computer readable storage medium or to an external computer or external storage device via a network, for example, the Internet, a local area network, a wide area network and/or a wireless network. The network may comprise copper transmission cables, optical transmission fibers, wireless transmission, routers, firewalls, switches, gateway computers and/or edge servers. A network adapter card or network interface in each computing/processing device receives computer readable program instructions from the network and forwards the computer readable program instructions for storage in a computer readable storage medium within the respective computing/processing device.
• Computer readable program instructions for carrying out operations of the present invention may be assembler instructions, instruction-set-architecture (ISA) instructions, machine instructions, machine dependent instructions, microcode, firmware instructions, state-setting data, or either source code or object code written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C++ or the like, and conventional procedural programming languages, such as the “C” programming language or similar programming languages. The computer readable program instructions may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet Service Provider). In some embodiments, electronic circuitry including, for example, programmable logic circuitry, field-programmable gate arrays (FPGA), or programmable logic arrays (PLA) may execute the computer readable program instructions by utilizing state information of the computer readable program instructions to personalize the electronic circuitry, in order to perform aspects of the present invention.
• Aspects of the present invention are described herein with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer readable program instructions.
• These computer readable program instructions may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions/acts specified in the flowchart and/or block diagram block or blocks. These computer readable program instructions may also be stored in a computer readable storage medium that can direct a computer, a programmable data processing apparatus, and/or other devices to function in a particular manner, such that the computer readable storage medium having instructions stored therein comprises an article of manufacture including instructions which implement aspects of the function/act specified in the flowchart and/or block diagram block or blocks.
• The computer readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable apparatus or other device to produce a computer implemented process, such that the instructions which execute on the computer, other programmable apparatus, or other device implement the functions/acts specified in the flowchart and/or block diagram block or blocks.
• The flowchart and block diagrams in the Figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of instructions, which comprises one or more executable instructions for implementing the specified logical function(s). In some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems that perform the specified functions or acts or carry out combinations of special purpose hardware and computer instructions.
• The computational components ofFIG. 1, including theuser computing devices106₁,106_i. . .106_nandinformation manager server100, may be implemented in one or more computer systems, such as thecomputer system1002 shown inFIG. 10. Computer system/server1002 may be described in the general context of computer system executable instructions, such as program modules, being executed by a computer system. Generally, program modules may include routines, programs, objects, components, logic, data structures, and so on that perform particular tasks or implement particular abstract data types. Computer system/server1002 may be practiced in distributed cloud computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed cloud computing environment, program modules may be located in both local and remote computer system storage media including memory storage devices.
• As shown inFIG. 11, the computer system/server1102 is shown in the form of a general-purpose computing device. The components of computer system/server1102 may include, but are not limited to, one or more processors orprocessing units1104, asystem memory1106, and abus1108 that couples various system components includingsystem memory1106 toprocessor1104.Bus1108 represents one or more of any of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, and a processor or local bus using any of a variety of bus architectures. By way of example, and not limitation, such architectures include Industry Standard Architecture (ISA) bus, Micro Channel Architecture (MCA) bus, Enhanced ISA (EISA) bus, Video Electronics Standards Association (VESA) local bus, and Peripheral Component Interconnects (PCI) bus.
• Computer system/server1102 typically includes a variety of computer system readable media. Such media may be any available media that is accessible by computer system/server1102, and it includes both volatile and non-volatile media, removable and non-removable media.
• System memory1106 can include computer system readable media in the form of volatile memory, such as random access memory (RAM)1110 and/orcache memory1112. Computer system/server1102 may further include other removable/non-removable, volatile/non-volatile computer system storage media. By way of example only,storage system1113 can be provided for reading from and writing to a non-removable, non-volatile magnetic media (not shown and typically called a “hard drive”). Although not shown, a magnetic disk drive for reading from and writing to a removable, non-volatile magnetic disk (e.g., a “floppy disk”), and an optical disk drive for reading from or writing to a removable, non-volatile optical disk such as a CD-ROM, DVD-ROM or other optical media can be provided. In such instances, each can be connected tobus1108 by one or more data media interfaces. As will be further depicted and described below,memory1106 may include at least one program product having a set (e.g., at least one) of program modules that are configured to carry out the functions of embodiments of the invention.
• Program/utility1114, having a set (at least one) ofprogram modules1116, may be stored inmemory1106 by way of example, and not limitation, as well as an operating system, one or more application programs, other program modules, and program data. Each of the operating system, one or more application programs, other program modules, and program data or some combination thereof, may include an implementation of a networking environment. The components of thecomputer1102 may be implemented asprogram modules1116 which generally carry out the functions and/or methodologies of embodiments of the invention as described herein. The systems ofFIG. 1 may be implemented in one ormore computer systems1102, where if they are implemented inmultiple computer systems1102, then the computer systems may communicate over a network.
• Computer system/server1102 may also communicate with one or moreexternal devices1118 such as a keyboard, a pointing device, adisplay1120, etc.; one or more devices that enable a user to interact with computer system/server1102; and/or any devices (e.g., network card, modem, etc.) that enable computer system/server1102 to communicate with one or more other computing devices. Such communication can occur via Input/Output (I/O) interfaces1122. Still yet, computer system/server1102 can communicate with one or more networks such as a local area network (LAN), a general wide area network (WAN), and/or a public network (e.g., the Internet) vianetwork adapter1124. As depicted,network adapter1124 communicates with the other components of computer system/server1102 viabus1108. It should be understood that although not shown, other hardware and/or software components could be used in conjunction with computer system/server1102. Examples, include, but are not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data archival storage systems, etc.
• The terms “an embodiment”, “embodiment”, “embodiments”, “the embodiment”, “the embodiments”, “one or more embodiments”, “some embodiments”, and “one embodiment” mean “one or more (but not all) embodiments of the present invention(s)” unless expressly specified otherwise.
• The terms “including”, “comprising”, “having” and variations thereof mean “including but not limited to”, unless expressly specified otherwise.
• The enumerated listing of items does not imply that any or all of the items are mutually exclusive, unless expressly specified otherwise.
• The terms “a”, “an” and “the” mean “one or more”, unless expressly specified otherwise.
• Devices that are in communication with each other need not be in continuous communication with each other, unless expressly specified otherwise. In addition, devices that are in communication with each other may communicate directly or indirectly through one or more intermediaries.
• A description of an embodiment with several components in communication with each other does not imply that all such components are required. On the contrary a variety of optional components are described to illustrate the wide variety of possible embodiments of the present invention.
• When a single device or article is described herein, it will be readily apparent that more than one device/article (whether or not they cooperate) may be used in place of a single device/article. Similarly, where more than one device or article is described herein (whether or not they cooperate), it will be readily apparent that a single device/article may be used in place of the more than one device or article or a different number of devices/articles may be used instead of the shown number of devices or programs. The functionality and/or the features of a device may be alternatively embodied by one or more other devices which are not explicitly described as having such functionality/features. Thus, other embodiments of the present invention need not include the device itself.
• The foregoing description of various embodiments of the invention has been presented for the purposes of illustration and description. It is not intended to be exhaustive or to limit the invention to the precise form disclosed. Many modifications and variations are possible in light of the above teaching. It is intended that the scope of the invention be limited not by this detailed description, but rather by the claims appended hereto. The above specification, examples and data provide a complete description of the manufacture and use of the composition of the invention. Since many embodiments of the invention can be made without departing from the spirit and scope of the invention, the invention resides in the claims herein after appended.

Claims (21)

1-22. (canceled)

23. A computer program product for determining an order in which to present items in a document for review in a computer user interface, wherein the computer program product comprises a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor to communicate with a server and cause operations, the operations comprising:

detecting users interactions with items included in reviewed documents, wherein the users operate Input/Output (I/O) devices to interact with the items in the reviewed documents presented in computer user interfaces;

determining context attribute values for context attributes related to the users interactions with the items;

generating interaction information for the items with which the users interacted, wherein an instance of interaction information indicates the item and a determined context attribute value related to a user interaction with the item;

determining a context attribute value for a context attribute for a selected document having items;

determining interaction information for items in the selected document having a context attribute value matching the determined context attribute value of the selected document;

using the determined interaction information for the items in the selected document to determine an order in which to present the items in the selected document; and

presenting in a computer user interface the items in the selected document in the order for a reviewing user to consider.

24. The computer program product ofclaim 23, wherein the operations further comprise:

indicating, in the interaction information for the items, frequencies with which the users have interacted with the items in the reviewed documents, wherein the using the determined interaction information for the items to determine the order comprises ordering the items based on the frequencies in the determined interaction information for the items.

25. The computer program product ofclaim 24, wherein the operations further comprise:

determining cumulative interaction frequencies for the items in the selected document based on the frequencies in the determined interaction information for the items, wherein the cumulative interaction frequency for a given item comprises a sum of the frequencies in the determined interaction information for the given item, and wherein the order is based on the cumulative interaction frequencies of the items.

26. The computer program product ofclaim 23, wherein the context attribute values comprise one of a plurality of context attribute values that are members of a set of context attribute values comprising: a user attribute value for an attribute of a user when interacting with an item; a question subject attribute value indicating a subject of an interrogative sentence including an item interacted with by a user; an item specific attribute value specific to a context of the item; a location attribute value of where the user was located when interacting with the item; a time attribute value comprising a time period when the user was interacting with the item; and a user cohort value comprising one of a plurality of user cohort values of groups of users.

27. The computer program product ofclaim 26, wherein users comprise medical professionals, the subject of the interrogative sentence including the items comprises a patient, the items are related to medical treatment of patients, documents comprise patient medical records, wherein the location attribute value includes a type of medical facility where the user is located, wherein the question subject attribute value comprises at least one medical condition of the subject, and wherein the interrogative sentence including the items concerns gathering medical information related to the patients that are subjects of the interrogative sentence.

28. The computer program product ofclaim 23, wherein the presenting in the computer user interface the items in the selected document in the order comprises:

directing the reviewing user in the computer user interface to an item in the document not yet considered having a highest order of the items not yet considered to receive user modifications to content for the item;

saving user modifications to the content for the item in the selected document;

receiving user selection to move to a next item in the document; and

performing additional iterations of the directing the reviewing user to an item having a highest order of the items not yet considered and saving user modifications to the content for the item in response to receiving user selections to move to next items in the document.

29. The computer program product ofclaim 23, wherein the operations further comprise:

determining a profile value for the reviewing user from user profile information for the reviewing user, wherein the determining the interaction information for the items in the selected document further comprises determining interaction information also having a profile attribute value matching the determined profile value for the reviewing user.

30. A system for determining an order in which to present items in a document for review in communication with a server, comprising:

a processor; and

a computer readable storage medium having program instructions that when executed by the processor to cause operations, the operations comprising:

detecting users interactions of users with items included in reviewed documents, wherein the users operate Input/Output (I/O) devices to interact with the items in the reviewed documents presented in computer user interfaces;

determining context attribute values for context attributes related to the users interactions with the items;

generating interaction information for the items with which the users interacted, wherein an instance of interaction information indicates the item and a determined context attribute value related to a user interaction with the item;

determining a context attribute value for a context attribute for a selected document having items;

determining interaction information for items in the selected document having a context attribute value matching the determined context attribute value of the selected document;

using the determined interaction information for the items in the selected document to determine an order in which to present the items in the selected document; and

presenting in a computer user interface the items in the selected document in the order for a reviewing user to consider.

31. The system ofclaim 30, wherein the operations further comprise:

indicating, in the interaction information for the items, frequencies with which the users have interacted with the items in the reviewed documents, wherein the using the determined interaction information for the items to determine the order comprises ordering the items based on the frequencies in the determined interaction information for the items.

32. The system ofclaim 31, wherein the operations further comprise:

determining cumulative interaction frequencies for the items in the selected document based on the frequencies in the determined interaction information for the items, wherein the cumulative interaction frequency for a given item comprises a sum of the frequencies in the determined interaction information for the given item, and wherein the order is based on the cumulative interaction frequencies of the items.

33. The system ofclaim 30, wherein the context attribute values comprise one of a plurality of context attribute values that are members of a set of context attribute values comprising: a user attribute value for an attribute of a user when interacting with an item; a question subject attribute value indicating a subject of an interrogative sentence including an item interacted with by a user; an item specific attribute value specific to a context of the item; a location attribute value of where the user was located when interacting with the item; a time attribute value comprising a time period when the user was interacting with the item; and a user cohort value comprising one of a plurality of user cohort values of groups of users.

34. The system ofclaim 33, wherein users comprise medical professionals, the subject of the interrogative sentence including the items comprises a patient, the items are related to medical treatment of patients, documents comprise patient medical records, wherein the location attribute value includes a type of medical facility where the user is located, wherein the question subject attribute value comprises at least one medical condition of the subject, and wherein the interrogative sentence including the items concerns gathering medical information related to the patients that are subjects of the interrogative sentence.

35. The system ofclaim 30, wherein the presenting in the computer user interface the items in the selected document in the order comprises:

directing the reviewing user in the computer user interface to an item in the document not yet considered having a highest order of the items not yet considered to receive user modifications to content for the item;

saving user modifications to the content for the item in the selected document;

receiving user selection to move to a next item in the document; and

performing additional iterations of the directing the reviewing user to an item having a highest order of the items not yet considered and saving user modifications to the content for the item in response to receiving user selections to move to next items in the document.

36. The system ofclaim 30, wherein the operations further comprise:

determining a profile value for the reviewing user from user profile information for the reviewing user, wherein the determining the interaction information for the items in the selected document further comprises determining interaction information also having a profile attribute value matching the determined profile value for the reviewing user.

37. A method for determining an order in which to present items in a document for review in a computer user interface, comprising:

detecting users interactions

with items included in reviewed documents, wherein the users operate Input/Output (I/O) devices to interact with the items in the reviewed documents presented in computer user interfaces;

determining context attribute values for context attributes related to the users interactions with the items;

generating interaction information for the items with which the users interacted, wherein an instance of interaction information indicates the item and a determined context attribute value related to a user interaction with the item;

determining a context attribute value for a context attribute for a selected document having items;

determining interaction information for items in the selected document having a context attribute value matching the determined context attribute value of the selected document;

using the determined interaction information for the items in the selected document to determine an order in which to present the items in the selected document; and

presenting in a computer user interface the items in the selected document in the order for a reviewing user to consider.

38. The method ofclaim 37, further comprising:

indicating, in the interaction information for the items, frequencies with which the users have interacted with the items in the reviewed documents, wherein the using the determined interaction information for the items to determine the order comprises ordering the items based on the frequencies in the determined interaction information for the items.

39. The method ofclaim 38, further comprising:

determining cumulative interaction frequencies for the items in the selected document based on the frequencies in the determined interaction information for the items, wherein the cumulative interaction frequency for a given item comprises a sum of the frequencies in the determined interaction information for the given item, and wherein the order is based on the cumulative interaction frequencies of the items.

40. The method ofclaim 37, wherein the context attribute values comprise one of a plurality of context attribute values that are members of a set of context attribute values comprising: a user attribute value for an attribute of a user when interacting with an item; a question subject attribute value indicating a subject of an interrogative sentence including an item interacted with by a user; an item specific attribute value specific to a context of the item; a location attribute value of where the user was located when interacting with the item; a time attribute value comprising a time period when the user was interacting with the item; and a user cohort value comprising one of a plurality of user cohort values of groups of users.

41. The method ofclaim 40, wherein users comprise medical professionals, the subject of the interrogative sentence including the items comprises a patient, the items are related to medical treatment of patients, documents comprise patient medical records, wherein the location attribute value includes a type of medical facility where the user is located, wherein the question subject attribute value comprises at least one medical condition of the subject, and wherein the interrogative sentence including the items concerns gathering medical information related to the patients that are subjects of the interrogative sentence.

42. The method ofclaim 37, wherein the presenting in the computer user interface the items in the selected document in the order comprises:

directing the reviewing user in the computer user interface to an item in the document not yet considered having a highest order of the items not yet considered to receive user modifications to content for the item;

saving user modifications to the content for the item in the selected document;

receiving user selection to move to a next item in the document; and

performing additional iterations of the directing the reviewing user to an item having a highest order of the items not yet considered and saving user modifications to the content for the item in response to receiving user selections to move to next items in the document.

Images