US20160140216A1 - Adjusting Fact-Based Answers to Consider Outcomes - Google Patents
Adjusting Fact-Based Answers to Consider Outcomes Download PDFInfo
- Publication number
- US20160140216A1 US20160140216A1 US14/547,367 US201414547367A US2016140216A1 US 20160140216 A1 US20160140216 A1 US 20160140216A1 US 201414547367 A US201414547367 A US 201414547367A US 2016140216 A1 US2016140216 A1 US 2016140216A1
- Authority
- US
- United States
- Prior art keywords
- question
- answer
- computer
- user
- readable storage
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Abandoned
Links
Images
Classifications
-
- G06F17/30657—
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/02—Knowledge representation; Symbolic representation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/23—Updating
- G06F16/2308—Concurrency control
- G06F16/2315—Optimistic concurrency control
- G06F16/2322—Optimistic concurrency control using timestamps
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/332—Query formulation
- G06F16/3329—Natural language query formulation or dialogue systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/903—Querying
- G06F16/9032—Query formulation
- G06F16/90332—Natural language query formulation or dialogue systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/951—Indexing; Web crawling techniques
-
- G06F17/30654—
-
- G06F17/30705—
-
- G06F17/30864—
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/04—Inference or reasoning models
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/3332—Query translation
- G06F16/3334—Selection or weighting of terms from queries, including natural language queries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3344—Query execution using natural language analysis
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/08—Speech classification or search
- G10L15/18—Speech classification or search using natural language modelling
- G10L15/1822—Parsing for meaning understanding
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/48—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use
- G10L25/51—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use for comparison or discrimination
- G10L25/63—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use for comparison or discrimination for estimating an emotional state
Definitions
- the present invention relates in general to the field of computers and similar technologies, and in particular to software utilized in this field. Still more particularly, it relates to a method, system and computer-usable medium for adjusting fact-based answers provided by a question/answer (QA) system.
- QA question/answer
- Cognitive question/answer (QA) systems such as the IBM WatsonTM system available from International Business Machines (IBMTM), process questions posed in natural language to determine answers and associated confidence scores based upon various corpora of knowledge.
- users submit one or more questions through an application's user interface (UI) or application programming interface (API) to the QA system.
- UI user interface
- API application programming interface
- the questions are processed to generate fact-based answers, which are then returned to the user.
- these fact-based answers may include information that could be used inappropriately.
- the inappropriate use of such information could lead to illegal or unfortunate activities that result in financial, physical or emotional harm to the user or others.
- a user may ask the question, “Where can I jump from a bridge?” If the user's intent is to know where they might experience bungee-jumping, then the location of such a bridge is appropriate. However, if the user is suicidal, then providing the location of such a bridge could lead to the user's death.
- a user may ask, “Where can I purchase an automatic pistol?” If the user's intent is to purchase a pistol for target practice, then returning information related to a fire arms dealer is appropriate. However, if the user is fixated on perpetrating a school shooting, then provision of the same information may contribute to others suffering possible harm, or even death. Furthermore, the provision of such information could lead to possible legal exposure and litigation for the QA system provider.
- a method, system and computer-usable medium are disclosed for adjusting fact-based answers provided by a question/answer (QA) system.
- the QA system is trained to identify a question that might lead to an undesirable outcome.
- the method of training the QA system may include manual, automated or learning approaches.
- a user submits a question to the QA system, where it is categorized into a question type.
- the QA system then processes the question to generate an answer.
- the QA system then generates an answer adjustment if it is determined that the question type and answer meet a predicted undesirable outcome.
- the answer adjustment may include a warning, a disclaimer, a recommendation, an alternative fact-based answer, a referral to an assistance service, or any combination thereof.
- the answer and the answer adjustment is provided to the user.
- a time-stamped record is generated and retained, which includes the question, the answer, the answer adjustment, and identification information associated with the user.
- FIG. 1 depicts an exemplary client computer in which the present invention may be implemented
- FIG. 2 is a simplified block diagram of an information handling system capable of performing computing operations
- FIG. 3 is a simplified block diagram of a fact-based answer adjustment system used to generate adjustments to fact-based answers.
- FIG. 4 is a generalized flowchart of the performance of operations to generate adjustments to fact-based answers.
- a method, system and computer-usable medium are disclosed for adjusting fact-based answers provided by a question/answer (QA) system.
- the present invention may be a system, a method, and/or a computer program product.
- selected aspects of the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment (including firmware, resident software, micro-code, etc.) or an embodiment combining software and/or hardware aspects that may all generally be referred to herein as a “circuit,” “module” or “system.”
- aspects of the present invention may take the form of computer program product embodied in 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 dynamic or static random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), a magnetic storage device, 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 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 or cluster of servers.
- 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).
- 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.
- 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.
- 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).
- the functions noted in the block may occur out of the order noted in the figures.
- 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.
- FIG. 1 depicts a schematic diagram of one illustrative embodiment of a question prioritization system 10 and question/answer (QA) system 100 connected to a computer network 140 .
- the QA system 100 includes a knowledge manager 104 that is connected to a knowledge base 106 and configured to provide question/answer (QA) generation functionality for one or more content users who submit across the network 140 to the QA system 100 .
- the prioritization system 10 may be connected to the computer network 140 to receive user questions, and may include a plurality of subsystems which interact with cognitive systems, like the knowledge manager 100 , to prioritize questions or requests being submitted to the knowledge manager 100 .
- the Named Entity subsystem 12 receives and processes each question 11 by using natural language (NL) processing to analyze each question and extract question topic information contained in the question, such as named entities, phrases, urgent terms, and/or other specified terms which are stored in one or more domain entity dictionaries 13 .
- NL natural language
- the domain dictionary 11 enables critical and urgent words (e.g., “threat level”) from different domains (e.g., “travel”) to be identified in each question based on their presence in the domain dictionary 11 .
- NLP Natural Language Processing
- NLP refers to the field of computer science, artificial intelligence, and linguistics concerned with the interactions between computers and human (natural) languages.
- NLP is related to the area of human—computer interaction and natural language understanding by computer systems that enable computer systems to derive meaning from human or natural language input.
- NLP can be used to derive meaning from a human-oriented question such as, “What is tallest mountain in North America?” and to identify specified terms, such as named entities, phrases, or urgent terms contained in the question.
- the process identifies key terms and attributes in the question and compares the identified terms to the stored terms in the domain dictionary 13 .
- the Question Priority Manager subsystem 14 performs additional processing on each question to extract question context information 15 A.
- the Question Priority Manager subsystem 14 may also extract server performance information 15 B for the question prioritization system 10 and/or QA system 100 .
- the extracted question context information 15 A may include data that identifies the user context and location when the question was submitted or received.
- the extracted question context information 15 A may include data that identifies the user who submitted the question (e.g., through login credentials), the device or computer which sent the question, the channel over which the question was submitted, the location of the user or device that sent the question, any special interest location indicator (e.g., hospital, public-safety answering point, etc.), or other context-related data for the question.
- any special interest location indicator e.g., hospital, public-safety answering point, etc.
- the Question Priority Manager subsystem 14 may also determine or extract selected server performance data 15 B for the processing of each question.
- the server performance information 15 B may include operational metric data relating to the available processing resources at the question prioritization system 10 and/or QA system 100 , such as operational or run-time data, CPU utilization data, available disk space data, bandwidth utilization data, etc.
- the Question Priority Manager subsystem 14 may identify the SLA or QoS processing requirements that apply to the question being analyzed, the history of analysis and feedback for the question or submitting user, and the like.
- the Question Priority Manager subsystem 14 is configured to populate feature values for the Priority Assignment Model 16 which provides a machine learning predictive model for generating a target priority values for the question, such as by using an artificial intelligence (AI) rule-based logic to determine and assign a question urgency value to each question for purposes of prioritizing the response processing of each question by the QA system 100 .
- AI artificial intelligence
- the Prioritization Manager subsystem 17 performs additional sort or rank processing to organize the received questions based on at least the associated target priority values such that high priority questions are put to the front of a prioritized question queue 18 for output as prioritized questions 19 .
- the highest priority question is placed at the front for delivery to the assigned QA system 100 .
- the prioritized questions 19 from the Prioritization Manager subsystem 17 that have a specified target priority value may be assigned to a specific pipeline (e.g., QA System 100 A) in the QA system cluster 100 .
- the Prioritization Manager subsystem 17 may use the question queue 18 as a message queue to provide an asynchronous communications protocol for delivering prioritized questions 19 to the QA system 100 such that the Prioritization Manager subsystem 17 and QA system 100 do not need to interact with a question queue 18 at the same time by storing prioritized questions in the question queue 18 until the QA system 100 retrieves them.
- a wider asynchronous network supports the passing of prioritized questions as messages between different computer systems 100 A, 100 B, connecting multiple applications and multiple operating systems. Messages can also be passed from queue to queue in order for a message to reach the ultimate desired recipient.
- An example of a commercial implementation of such messaging software is IBM's WebSphere MQ (previously MQ Series).
- the organizational function of the Prioritization Manager subsystem 17 may be configured to convert over-subscribing questions into asynchronous responses, even if they were asked in a synchronized fashion.
- the QA system 100 may include one or more QA system pipelines 100 A, 100 B, each of which includes a computing device 104 (comprising one or more processors and one or more memories, and potentially any other computing device elements generally known in the art including buses, storage devices, communication interfaces, and the like) for processing questions received over the network 140 from one or more users at computing devices (e.g., 110 , 120 , 130 ) connected over the network 140 for communication with each other and with other devices or components via one or more wired and/or wireless data communication links, where each communication link may comprise one or more of wires, routers, switches, transmitters, receivers, or the like.
- a computing device 104 comprising one or more processors and one or more memories, and potentially any other computing device elements generally known in the art including buses, storage devices, communication interfaces, and the like
- computing devices e.g., 110 , 120 , 130
- each communication link may comprise one or more of wires, routers, switches, transmitters, receivers, or the like.
- the QA system 100 and network 140 may enable question/answer (QA) generation functionality for one or more content users.
- QA system 100 may be used with components, systems, sub-systems, and/or devices other than those that are depicted herein.
- a prioritized question 19 is received and prioritized for processing to generate an answer 20 .
- prioritized questions 19 are dequeued from the shared question queue 18 , from which they are de-queued by the pipeline instances for processing in priority order rather than insertion order.
- the question queue 18 may be implemented based on a “priority heap” data structure.
- questions may be split into many subtasks which run concurrently. A single pipeline instance can process a number of questions concurrently, but only a certain number of subtasks.
- each QA system pipeline may include a prioritized queue (not shown) to manage the processing order of these subtasks, with the top-level priority corresponding to the time that the corresponding question started (earliest has highest priority).
- a prioritized queue (not shown) to manage the processing order of these subtasks, with the top-level priority corresponding to the time that the corresponding question started (earliest has highest priority).
- the external target priority values generated for each question by the Question Priority Manager subsystem 14 may take precedence or ranking priority over the question start time. In this way, more important or higher priority questions can “fast track” through the QA system pipeline if it is busy with already-running questions.
- the knowledge manager 104 may be configured to receive inputs from various sources.
- knowledge manager 104 may receive input from the question prioritization system 10 , network 140 , a knowledge base or corpus of electronic documents 106 or other data, a content creator 108 , content users, and other possible sources of input.
- some or all of the inputs to knowledge manager 104 may be routed through the network 140 and/or the question prioritization system 10 .
- the various computing devices e.g., 110 , 120 , 130 , 150 , 160 , 170
- on the network 140 may include access points for content creators and content users.
- Some of the computing devices may include devices for a database storing the corpus of data as the body of information used by the knowledge manager 104 to generate answers to cases.
- the network 140 may include local network connections and remote connections in various embodiments, such that knowledge manager 104 may operate in environments of any size, including local and global, e.g., the Internet.
- knowledge manager 104 serves as a front-end system that can make available a variety of knowledge extracted from or represented in documents, network-accessible sources and/or structured data sources. In this manner, some processes populate the knowledge manager with the knowledge manager also including input interfaces to receive knowledge requests and respond accordingly.
- the content creator creates content in a document 106 for use as part of a corpus of data with knowledge manager 104 .
- the document 106 may include any file, text, article, or source of data (e.g., scholarly articles, dictionary definitions, encyclopedia references, and the like) for use in knowledge manager 104 .
- Content users may access knowledge manager 104 via a network connection or an Internet connection to the network 140 , and may input questions to knowledge manager 104 that may be answered by the content in the corpus of data.
- the process can use a variety of conventions to query it from the knowledge manager. One convention is to send a well-formed question.
- Semantic content is content based on the relation between signifiers, such as words, phrases, signs, and symbols, and what they stand for, their denotation, or connotation.
- semantic content is content that interprets an expression, such as by using Natural Language (NL) Processing.
- the process sends well-formed questions (e.g., natural language questions, etc.) to the knowledge manager.
- Knowledge manager 104 may interpret the question and provide a response to the content user containing one or more answers to the question.
- knowledge manager 104 may provide a response to users in a ranked list of answers.
- QA system 100 may be the IBM WatsonTM QA system available from International Business Machines Corporation of Armonk, N.Y., which is augmented with the mechanisms of the illustrative embodiments described hereafter.
- the IBM WatsonTM knowledge manager system may receive an input question which it then parses to extract the major features of the question, that in turn are then used to formulate queries that are applied to the corpus of data. Based on the application of the queries to the corpus of data, a set of hypotheses, or candidate answers to the input question, are generated by looking across the corpus of data for portions of the corpus of data that have some potential for containing a valuable response to the input question.
- the IBM WatsonTM QA system then performs deep analysis on the language of the input prioritized question 19 and the language used in each of the portions of the corpus of data found during the application of the queries using a variety of reasoning algorithms.
- reasoning algorithms There may be hundreds or even thousands of reasoning algorithms applied, each of which performs different analysis, e.g., comparisons, and generates a score.
- some reasoning algorithms may look at the matching of terms and synonyms within the language of the input question and the found portions of the corpus of data.
- Other reasoning algorithms may look at temporal or spatial features in the language, while others may evaluate the source of the portion of the corpus of data and evaluate its veracity.
- the scores obtained from the various reasoning algorithms indicate the extent to which the potential response is inferred by the input question based on the specific area of focus of that reasoning algorithm. Each resulting score is then weighted against a statistical model.
- the statistical model captures how well the reasoning algorithm performed at establishing the inference between two similar passages for a particular domain during the training period of the IBM WatsonTM QA system. The statistical model may then be used to summarize a level of confidence that the IBM WatsonTM QA system has regarding the evidence that the potential response, i.e. candidate answer, is inferred by the question.
- the QA system 100 then generates an output response or answer 20 with the final answer and associated confidence and supporting evidence. More information about the IBM WatsonTM QA system may be obtained, for example, from the IBM Corporation website, IBM Redbooks, and the like. For example, information about the IBM WatsonTM QA system can be found in Yuan et al., “Watson and Healthcare,” IBM developerWorks, 2011 and “The Era of Cognitive Systems: An Inside Look at IBM Watson and How it Works” by Rob High, IBM Redbooks, 2012.
- Types of information processing systems that can utilize QA system 100 range from small handheld devices, such as handheld computer/mobile telephone 110 to large mainframe systems, such as mainframe computer 170 .
- handheld computer 110 include personal digital assistants (PDAs), personal entertainment devices, such as MP3 players, portable televisions, and compact disc players.
- PDAs personal digital assistants
- Other examples of information processing systems include pen, or tablet, computer 120 , laptop, or notebook, computer 130 , personal computer system 150 , and server 160 . As shown, the various information processing systems can be networked together using computer network 140 .
- Types of computer network 140 that can be used to interconnect the various information processing systems include Local Area Networks (LANs), Wireless Local Area Networks (WLANs), the Internet, the Public Switched Telephone Network (PSTN), other wireless networks, and any other network topology that can be used to interconnect the information processing systems.
- Many of the information processing systems include nonvolatile data stores, such as hard drives and/or nonvolatile memory. Some of the information processing systems may use separate nonvolatile data stores (e.g., server 160 utilizes nonvolatile data store 165 , and mainframe computer 170 utilizes nonvolatile data store 175 ).
- the nonvolatile data store can be a component that is external to the various information processing systems or can be internal to one of the information processing systems.
- FIG. 2 An illustrative example of an information processing system showing an exemplary processor and various components commonly accessed by the processor is shown in FIG. 2 .
- FIG. 2 illustrates an information processing system 202 , more particularly, a processor and common components, which is a simplified example of a computer system capable of performing the computing operations described herein.
- Information processing system 202 includes a processor unit 204 that is coupled to a system bus 206 .
- a video adapter 208 which controls a display 210 , is also coupled to system bus 206 .
- System bus 206 is coupled via a bus bridge 212 to an Input/Output (I/O) bus 214 .
- An I/O interface 216 is coupled to I/O bus 214 .
- I/O Input/Output
- the I/O interface 216 affords communication with various I/O devices, including a keyboard 218 , a mouse 220 , a Compact Disk-Read Only Memory (CD-ROM) drive 222 , a floppy disk drive 224 , and a flash drive memory 226 .
- the format of the ports connected to I/O interface 216 may be any known to those skilled in the art of computer architecture, including but not limited to Universal Serial Bus (USB) ports.
- USB Universal Serial Bus
- the information processing system 202 is able to communicate with a service provider server 252 via a network 228 using a network interface 230 , which is coupled to system bus 206 .
- Network 228 may be an external network such as the Internet, or an internal network such as an Ethernet Network or a Virtual Private Network (VPN).
- client computer 202 is able to use the present invention to access service provider server 252 .
- VPN Virtual Private Network
- a hard drive interface 232 is also coupled to system bus 206 .
- Hard drive interface 232 interfaces with a hard drive 234 .
- hard drive 234 populates a system memory 236 , which is also coupled to system bus 206 .
- Data that populates system memory 236 includes the information processing system's 202 operating system (OS) 238 and software programs 244 .
- OS operating system
- OS 238 includes a shell 240 for providing transparent user access to resources such as software programs 244 .
- shell 240 is a program that provides an interpreter and an interface between the user and the operating system. More specifically, shell 240 executes commands that are entered into a command line user interface or from a file.
- shell 240 (as it is called in UNIX®), also called a command processor in Windows®, is generally the highest level of the operating system software hierarchy and serves as a command interpreter.
- the shell provides a system prompt, interprets commands entered by keyboard, mouse, or other user input media, and sends the interpreted command(s) to the appropriate lower levels of the operating system (e.g., a kernel 242 ) for processing.
- a kernel 242 the appropriate lower levels of the operating system for processing.
- shell 240 generally is a text-based, line-oriented user interface
- the present invention can also support other user interface modes, such as graphical, voice, gestural, etc.
- OS 238 also includes kernel 242 , which includes lower levels of functionality for OS 238 , including essential services required by other parts of OS 238 and software programs 244 , including memory management, process and task management, disk management, and mouse and keyboard management.
- Software programs 244 may include a browser 246 and email client 248 .
- Browser 246 includes program modules and instructions enabling a World Wide Web (WWW) client (i.e., information processing system 202 ) to send and receive network messages to the Internet using HyperText Transfer Protocol (HTTP) messaging, thus enabling communication with service provider server 252 .
- WWW World Wide Web
- software programs 244 may also include a fact-based answer adjustment system 250 .
- the fact-based answer adjustment system 250 includes code for implementing the processes described hereinbelow.
- information processing system 202 is able to download the fact-based answer adjustment system 250 from a service provider server 252 .
- the hardware elements depicted in the information processing system 202 are not intended to be exhaustive, but rather are representative to highlight components used by the present invention.
- the information processing system 202 may include alternate memory storage devices such as magnetic cassettes, Digital Versatile Disks (DVDs), Bernoulli cartridges, and the like. These and other variations are intended to be within the spirit, scope and intent of the present invention.
- FIG. 3 is a simplified block diagram of a fact-based answer adjustment system implemented in accordance with an embodiment of the invention to generate adjustments to fact-based answers.
- a question/answer (QA) system 100 is implemented to generate candidate fact-based answers 318 in response to receiving a question 316 from a user 302 .
- the QA system 100 parses the question 316 to extract the major features of the question 316 . In turn, these features are then used to formulate queries that are applied to the corpus of data, such as a knowledge base 322 . Based upon the application of the queries to the corpus of data, a set of hypotheses, or candidate fact-based answers 318 to the question 316 , are generated.
- the QA system 100 is trained to identify questions 316 that may lead to undesirable outcomes.
- the QA system 100 is manually trained.
- a user manually flags a set of questions.
- training approaches familiar to skilled practitioners of the art are implemented with the QA system 100 such that it learns that questions 316 that have high term or sentence similarity should be flagged.
- a term refers to words and compound words whose meaning is dependent upon the context in which they are used.
- term similarity broadly refers to terms that are similar to terms stored in a repository of data, such as the knowledge base 322 .
- sentence similarity broadly refers to sentences that are similar to sentences stored in a repository of data, such as the knowledge base 322 .
- automated keyword training is implemented to automate the training of the QA system 100 .
- a user creates a list of keywords, and questions containing those keywords and related synonyms, that should be flagged.
- the QA system 100 observes a question 316 submitted by a user 302 , along with future events associated with the user 302 (e.g., via social media interactions) that are related to the question 316 .
- two users 302 may ask the same question 316 , “Where can I jump off a bridge?”
- One user 302 posts photographs on a social media site of themselves base jumping off a bridge, while the other user 302 is mentioned in an obituary that mentions a bridge.
- the QA system 100 maintains a record of questions 316 asked by a user 302 while also ingesting the user's 302 posts to various social media venues.
- the time and date of the user's questions 316 are indexed to the time and date of the user's social media postings.
- feature extraction operations are performed to detect various image features within photographs that the user 302 may have posted.
- the feature extraction operations may detect an image of a bridge and an image of a parachute in a photograph posted by the user.
- keywords such as “bridge” and “parachute” may be associated with the question 316 as inferred input.
- natural language processing (NLP) of text associated with the posted photograph may also reference the user's 302 desire to base jump off of a bridge, as shown in the posted photograph.
- NLP natural language processing
- keywords extracted from various questions 316 and social media posts associated with the user 302 are processed to identify synonym and concept similarities, which are in turn used to generate a keyword relevance score.
- keywords extracted from various questions 316 and social media posts associated with the user 302 are processed with their respective time and date stamps to generate a decaying relevance score.
- a user 302 makes two social media posts that are associated with a question 316 submitted by the user 302 to the QA system 100 . The first post takes place a week after the question 316 was submitted and the second post takes place a year after the question 316 was submitted. In this example, the post made a week after the question 316 was submitted has a higher relevance score than the post made a year after the question 316 was submitted.
- a user 302 asks the question 316 , “Where can I jump off a bridge?” on day 1 .
- the user 302 has posted a photograph of themselves fishing on a bridge.
- the caption for the photograph is, “A bad day of fishing is better than a good day at work.”
- the photograph and its associated caption are processed to generate keywords including “bridge,” “fishing,” and “work.”
- a second photograph is then posted by the user 302 on the same day, showing the user 302 jumping off the bridge to swim.
- the second photograph is processed, as described in greater detail herein, to generate inferred keywords that include “bridge,” “jumping,” and “swim.”
- a third photograph is posted by the user 302 , showing an aerial view of a river with a caption saying, “base jumping off of Royal Gorge Bridge.”
- the third photograph and its associated caption are processed to generate keywords that include “bridge” and “base jumping.
- the user 302 posts a fourth photograph of them and a friend, with a bridge in the background, along with an associated caption saying, “Just hanging out.”
- the photograph and its associated caption are then processed to generate keywords and phrases that include “bridge” and “hanging out.”
- the resulting keyword relevance score indicates that bridges are related to “fishing,” “base jumping,” and “hanging out.”
- the resulting decaying relevance score would give a higher association to fishing and jumping. From the foregoing, it would not be assumed that hanging out with friends at bridges is related to jumping off of them since the time intervals between the social media posts are so far apart.
- the QA system 100 uses a fact-based answer adjustment system 250 to process candidate fact-based answers to identify potentially undesirable outcomes or flagged conditions associated with the question 316 .
- the processing analyzes the candidate fact-based answers with reference to potentially undesirable outcome criteria identifying the potentially undesirable outcomes and flagged conditions.
- a potentially undesirable outcome broadly refers to a potentially illegal, risky or harmful result arising from the use of a fact-based answer provided by the QA system 100 .
- misuse of a fact-based answer 318 provided by the QA system 100 may result in the incurrence of financial risk, physical damage, death, pain, or legal liability.
- flagged conditions broadly refer to potentially undesirable implied or inferred behaviors, activities, events, intents or actions associated with the question 316 .
- flagged conditions broadly refer to behaviors which could lead to a potentially undesirable outcome.
- the question 316 may contain a reference to base-jumping, which is a risky activity that may result in physical harm, death, legal liability, or a combination thereof.
- base-jumping may be referenced as a flagged condition in various embodiments by the QA system 100 , the fact-based answer adjustment system 250 , or both. Skilled practitioners of the art will realize that many such examples are possible and the foregoing is not intended to limit the spirit, scope or intent of the invention.
- the QA system 100 evaluates the question 316 in various embodiments to consider not just a candidate fact-based answer 318 , but the possible kinds of outcomes associated with the question 316 .
- fact-based answer adjustments 320 to the fact-based answers 318 are generated by the fact-based answer adjustment system 250 when undesirable outcomes or flagged conditions associated with the question 316 are identified.
- the fact-base answer adjustments 320 may include a warning, a disclaimer, a recommendation, an alternative fact-based answer 318 , a referral to an assistance service, or any combination thereof.
- the QA system 100 may receive the question 316 , “Where can I jump off of a bridge?”
- suicide may be identified as a potentially undesirable outcome
- base jumping and bungee jumping may be identified as flagged conditions.
- the fact-based answer adjustments 320 may include a suicide prevention link, a warning about the dangers associated with bungee jumping, a legal disclaimer that base jumping from many bridges may be illegal, or a combination thereof.
- a user 302 uses a user device 304 to submit a question to the question/answer (QA) system 100 .
- a user device 304 refers to an information processing system such as a personal computer, a laptop computer, a tablet computer, a personal digital assistant (PDA), a smart phone, a mobile telephone, or other device that is capable of communicating and processing data.
- the user device 304 is used to exchange information between the user 302 and the QA system 100 through the use of a network 228 .
- the network 228 may be a public network, such as the Internet, a physical private network, a virtual private network (VPN), or any combination thereof.
- the network 228 may be a wireless network, including a personal area network (PAN), based on technologies such as Bluetooth or Ultra Wideband (UWB).
- the wireless network may include a wireless local area network (WLAN), based on variations of the IEEE 802.11 specification, often referred to as WiFi.
- the wireless network may include a wireless wide area network (WWAN) based on an industry standard including two and a half generation (2.5G) wireless technologies such as global system for mobile communications (GPRS) and enhanced data rates for GSM evolution (EDGE).
- WWANs based on existing third generation (3G) wireless technologies including universal mobile telecommunications system (UMTS) and wideband code division multiple access (W-CDMA).
- 3G third generation
- UMTS universal mobile telecommunications system
- W-CDMA wideband code division multiple access
- inventions may include the implementation of other 3G technologies, including evolution-data optimized (EVDO), IEEE 802.16 (WiMAX), wireless broadband (WiBro), high-speed downlink packet access (HSDPA), high-speed uplink packet access (HSUPA), and emerging fourth generation (4G) wireless technologies.
- EVDO evolution-data optimized
- WiMAX IEEE 802.16
- WiBro wireless broadband
- HSDPA high-speed downlink packet access
- HSUPA high-speed uplink packet access
- 4G fourth generation
- operations to generate adjustments to fact-based answers generated by a QA system 100 are begun by receiving user identification information 314 associated with a user 302 .
- the type of user identification information 314 and the method by which it is received, is a matter of design choice.
- the user identification information 314 may include a user identifier (ID) and password associated with the user 302 .
- the user identification information 314 may include the Internet Protocol (IP) address of a user device 304 being used by the user 302 .
- IP Internet Protocol
- the user identification information 314 is entered by the user 302 into a User Interface (UI) window 312 associated with a user device 304 being used by the user 302 .
- UI User Interface
- a question generated by the user 302 is then received by the QA system 100 for processing.
- the QA system 100 may receive the question as a result of the user 302 entering the question 316 within a UI window 312 of a user device 304 being used by the user 302 .
- the QA system 100 may receive the question 316 through an application programming interface (API) associated with a user device 304 used by the user 302 .
- API application programming interface
- the question 316 received from the user 302 is then processed by the QA system 100 to generate one or more candidate answers, which are in turn processed to identify potentially undesirable outcomes and flagged conditions associated with the question 316 .
- the QA system 100 uses a fact-based answer adjustment system 250 to process the candidate questions to identify the potentially undesirable outcomes and flagged conditions. If no potentially undesirable outcomes or flagged conditions were identified, then one or more fact-based answers 318 to the question are generated provided to the user 302 . Otherwise, one or more fact-based answers 318 to the question 316 , and associated fact-based answer adjustments 320 , are generated and provided to the user 302 .
- the associated fact-base answer adjustments 320 may include a warning, a disclaimer, a recommendation, an alternative fact-based answer, a referral to an assistance service, or any combination thereof.
- the fact-based answer adjustments 320 are generated by the fact-based answer adjustment system 250 using a repository of answer adjustment data 324 .
- the one or more fact-based answers 318 , and any associated fact-based answer adjustments 320 are provided to the user 302 within a UI window 312 of a user device 304 used by the user 302 .
- a time-stamped record of everything that was provided to the user 302 in response to receiving the question 316 , as well as associated user identification information 314 is generated and retained.
- the resulting time-stamped record is stored in a repository of time-stamped answer data 326 .
- FIG. 4 is a generalized flowchart of operations performed in accordance with an embodiment of the invention to generate adjustments to fact-based answers.
- operations to generate adjustments to fact-based answers are begun in step 402 , followed by a question/answer (QA) system, described in greater detail herein, receiving user identification information in step 404 .
- the type of user identification information, and the method by which it is received, is a matter of design choice.
- the user identification information may include a user identifier (ID) and password associated with the user.
- the identification information may include the Internet Protocol (IP) address of a system being used by the user.
- IP Internet Protocol
- a question generated by the user is then received by the QA system for processing in step 406 .
- the QA system may receive the question as a result of a user entering the question within a user interface (UI) window of a user device.
- the QA system may receive the question through an application programming interface (API) associated with a user device.
- API application programming interface
- the question received from the user is then processed by the QA system in step 408 to generate one or more candidate answers, which are in turn processed in step 410 to identify potentially undesirable outcomes and flagged conditions, described in greater detail herein, associated with the question.
- a determination is then made in step 412 whether any potentially undesirable outcomes or flagged conditions were identified in step 410 . If not, then one or more fact-based answers to the question are generated in step 414 and provided to the user. Otherwise, one or more fact-based answers to the question, and associated fact-based answer adjustments, are generated in step 416 and provided to the user.
- the associated fact-base answer adjustments may include a warning, a disclaimer, a recommendation, an alternative fact-based answer, a referral to an assistance service, or any combination thereof.
- step 414 Thereafter, or after operations performed in step 414 have been completed, a time-stamped record of everything that was provided to the user in response to receiving the question, as well as associated user identification information, is generated and retained in step 416 .
- a determination is then made in step 420 whether another question is received from the user for processing by the QA system. If so, then the process is continued, proceeding with step 406 . Otherwise, operations to generate adjustments to fact-based answers are ended in step 422 .
Abstract
Description
- 1. Field of the Invention
- The present invention relates in general to the field of computers and similar technologies, and in particular to software utilized in this field. Still more particularly, it relates to a method, system and computer-usable medium for adjusting fact-based answers provided by a question/answer (QA) system.
- 2. Description of the Related Art
- Cognitive question/answer (QA) systems, such as the IBM Watson™ system available from International Business Machines (IBM™), process questions posed in natural language to determine answers and associated confidence scores based upon various corpora of knowledge. In operation, users submit one or more questions through an application's user interface (UI) or application programming interface (API) to the QA system. In turn, the questions are processed to generate fact-based answers, which are then returned to the user.
- At times, these fact-based answers may include information that could be used inappropriately. In particular, the inappropriate use of such information could lead to illegal or unfortunate activities that result in financial, physical or emotional harm to the user or others. For example, a user may ask the question, “Where can I jump from a bridge?” If the user's intent is to know where they might experience bungee-jumping, then the location of such a bridge is appropriate. However, if the user is suicidal, then providing the location of such a bridge could lead to the user's death.
- As another example, a user may ask, “Where can I purchase an automatic pistol?” If the user's intent is to purchase a pistol for target practice, then returning information related to a fire arms dealer is appropriate. However, if the user is fixated on perpetrating a school shooting, then provision of the same information may contribute to others suffering possible harm, or even death. Furthermore, the provision of such information could lead to possible legal exposure and litigation for the QA system provider.
- A method, system and computer-usable medium are disclosed for adjusting fact-based answers provided by a question/answer (QA) system. In various embodiments, the QA system is trained to identify a question that might lead to an undesirable outcome. In certain embodiments, the method of training the QA system may include manual, automated or learning approaches.
- In various embodiments, a user submits a question to the QA system, where it is categorized into a question type. The QA system then processes the question to generate an answer. The QA system then generates an answer adjustment if it is determined that the question type and answer meet a predicted undesirable outcome. In certain embodiments, the answer adjustment may include a warning, a disclaimer, a recommendation, an alternative fact-based answer, a referral to an assistance service, or any combination thereof. In various embodiments, the answer and the answer adjustment is provided to the user. In certain embodiments, a time-stamped record is generated and retained, which includes the question, the answer, the answer adjustment, and identification information associated with the user.
- The present invention may be better understood, and its numerous objects, features and advantages made apparent to those skilled in the art by referencing the accompanying drawings. The use of the same reference number throughout the several figures designates a like or similar element.
-
FIG. 1 depicts an exemplary client computer in which the present invention may be implemented; -
FIG. 2 is a simplified block diagram of an information handling system capable of performing computing operations; -
FIG. 3 is a simplified block diagram of a fact-based answer adjustment system used to generate adjustments to fact-based answers; and -
FIG. 4 is a generalized flowchart of the performance of operations to generate adjustments to fact-based answers. - A method, system and computer-usable medium are disclosed for adjusting fact-based answers provided by a question/answer (QA) system. The present invention may be a system, a method, and/or a computer program product. In addition, selected aspects of the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment (including firmware, resident software, micro-code, etc.) or an embodiment combining software and/or hardware aspects that may all generally be referred to herein as a “circuit,” “module” or “system.” Furthermore, aspects of the present invention may take the form of computer program product embodied in 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 dynamic or static random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), a magnetic storage device, 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 or cluster of servers. 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.
-
FIG. 1 depicts a schematic diagram of one illustrative embodiment of aquestion prioritization system 10 and question/answer (QA)system 100 connected to acomputer network 140. TheQA system 100 includes aknowledge manager 104 that is connected to aknowledge base 106 and configured to provide question/answer (QA) generation functionality for one or more content users who submit across thenetwork 140 to theQA system 100. To assist with efficient sorting and presentation of questions to theQA system 100, theprioritization system 10 may be connected to thecomputer network 140 to receive user questions, and may include a plurality of subsystems which interact with cognitive systems, like theknowledge manager 100, to prioritize questions or requests being submitted to theknowledge manager 100. - The Named
Entity subsystem 12 receives and processes eachquestion 11 by using natural language (NL) processing to analyze each question and extract question topic information contained in the question, such as named entities, phrases, urgent terms, and/or other specified terms which are stored in one or more domain entity dictionaries 13. By leveraging a plurality of pluggable domain dictionaries relating to different domains or areas (e.g., travel, healthcare, electronics, game shows, financial services), thedomain dictionary 11 enables critical and urgent words (e.g., “threat level”) from different domains (e.g., “travel”) to be identified in each question based on their presence in thedomain dictionary 11. To this end, the NamedEntity subsystem 12 may use a Natural Language Processing (NLP) routine to identify the question topic information in each question. As used herein, “NLP” refers to the field of computer science, artificial intelligence, and linguistics concerned with the interactions between computers and human (natural) languages. In this context, NLP is related to the area of human—computer interaction and natural language understanding by computer systems that enable computer systems to derive meaning from human or natural language input. For example, NLP can be used to derive meaning from a human-oriented question such as, “What is tallest mountain in North America?” and to identify specified terms, such as named entities, phrases, or urgent terms contained in the question. The process identifies key terms and attributes in the question and compares the identified terms to the stored terms in thedomain dictionary 13. - The Question
Priority Manager subsystem 14 performs additional processing on each question to extractquestion context information 15A. In addition or in the alternative, the QuestionPriority Manager subsystem 14 may also extractserver performance information 15B for thequestion prioritization system 10 and/orQA system 100. In selected embodiments, the extractedquestion context information 15A may include data that identifies the user context and location when the question was submitted or received. For example, the extractedquestion context information 15A may include data that identifies the user who submitted the question (e.g., through login credentials), the device or computer which sent the question, the channel over which the question was submitted, the location of the user or device that sent the question, any special interest location indicator (e.g., hospital, public-safety answering point, etc.), or other context-related data for the question. The QuestionPriority Manager subsystem 14 may also determine or extract selectedserver performance data 15B for the processing of each question. In selected embodiments, theserver performance information 15B may include operational metric data relating to the available processing resources at thequestion prioritization system 10 and/orQA system 100, such as operational or run-time data, CPU utilization data, available disk space data, bandwidth utilization data, etc. As part of the extractedinformation 15A/B, the QuestionPriority Manager subsystem 14 may identify the SLA or QoS processing requirements that apply to the question being analyzed, the history of analysis and feedback for the question or submitting user, and the like. Using the question topic information and extracted question context and/or server performance information, the QuestionPriority Manager subsystem 14 is configured to populate feature values for thePriority Assignment Model 16 which provides a machine learning predictive model for generating a target priority values for the question, such as by using an artificial intelligence (AI) rule-based logic to determine and assign a question urgency value to each question for purposes of prioritizing the response processing of each question by theQA system 100. - The
Prioritization Manager subsystem 17 performs additional sort or rank processing to organize the received questions based on at least the associated target priority values such that high priority questions are put to the front of a prioritizedquestion queue 18 for output as prioritizedquestions 19. In thequestion queue 18 of thePrioritization Manager subsystem 17, the highest priority question is placed at the front for delivery to the assignedQA system 100. In selected embodiments, the prioritizedquestions 19 from thePrioritization Manager subsystem 17 that have a specified target priority value may be assigned to a specific pipeline (e.g.,QA System 100A) in theQA system cluster 100. As will be appreciated, thePrioritization Manager subsystem 17 may use thequestion queue 18 as a message queue to provide an asynchronous communications protocol for delivering prioritizedquestions 19 to theQA system 100 such that thePrioritization Manager subsystem 17 andQA system 100 do not need to interact with aquestion queue 18 at the same time by storing prioritized questions in thequestion queue 18 until theQA system 100 retrieves them. In this way, a wider asynchronous network supports the passing of prioritized questions as messages betweendifferent computer systems Prioritization Manager subsystem 17 may be configured to convert over-subscribing questions into asynchronous responses, even if they were asked in a synchronized fashion. - The
QA system 100 may include one or moreQA system pipelines network 140 from one or more users at computing devices (e.g., 110, 120, 130) connected over thenetwork 140 for communication with each other and with other devices or components via one or more wired and/or wireless data communication links, where each communication link may comprise one or more of wires, routers, switches, transmitters, receivers, or the like. In this networked arrangement, theQA system 100 andnetwork 140 may enable question/answer (QA) generation functionality for one or more content users. Other embodiments ofQA system 100 may be used with components, systems, sub-systems, and/or devices other than those that are depicted herein. - In each
QA system pipeline question 19 is received and prioritized for processing to generate ananswer 20. In sequence, prioritizedquestions 19 are dequeued from the sharedquestion queue 18, from which they are de-queued by the pipeline instances for processing in priority order rather than insertion order. In selected embodiments, thequestion queue 18 may be implemented based on a “priority heap” data structure. During processing within a QA system pipeline (e.g., 100A), questions may be split into many subtasks which run concurrently. A single pipeline instance can process a number of questions concurrently, but only a certain number of subtasks. In addition, each QA system pipeline may include a prioritized queue (not shown) to manage the processing order of these subtasks, with the top-level priority corresponding to the time that the corresponding question started (earliest has highest priority). However, it will be appreciated that such internal prioritization within each QA system pipeline may be augmented by the external target priority values generated for each question by the QuestionPriority Manager subsystem 14 to take precedence or ranking priority over the question start time. In this way, more important or higher priority questions can “fast track” through the QA system pipeline if it is busy with already-running questions. - In the
QA system 100, theknowledge manager 104 may be configured to receive inputs from various sources. For example,knowledge manager 104 may receive input from thequestion prioritization system 10,network 140, a knowledge base or corpus ofelectronic documents 106 or other data, acontent creator 108, content users, and other possible sources of input. In selected embodiments, some or all of the inputs toknowledge manager 104 may be routed through thenetwork 140 and/or thequestion prioritization system 10. The various computing devices (e.g., 110, 120, 130, 150, 160, 170) on thenetwork 140 may include access points for content creators and content users. Some of the computing devices may include devices for a database storing the corpus of data as the body of information used by theknowledge manager 104 to generate answers to cases. Thenetwork 140 may include local network connections and remote connections in various embodiments, such thatknowledge manager 104 may operate in environments of any size, including local and global, e.g., the Internet. Additionally,knowledge manager 104 serves as a front-end system that can make available a variety of knowledge extracted from or represented in documents, network-accessible sources and/or structured data sources. In this manner, some processes populate the knowledge manager with the knowledge manager also including input interfaces to receive knowledge requests and respond accordingly. - In one embodiment, the content creator creates content in a
document 106 for use as part of a corpus of data withknowledge manager 104. Thedocument 106 may include any file, text, article, or source of data (e.g., scholarly articles, dictionary definitions, encyclopedia references, and the like) for use inknowledge manager 104. Content users may accessknowledge manager 104 via a network connection or an Internet connection to thenetwork 140, and may input questions toknowledge manager 104 that may be answered by the content in the corpus of data. As further described below, when a process evaluates a given section of a document for semantic content, the process can use a variety of conventions to query it from the knowledge manager. One convention is to send a well-formed question. Semantic content is content based on the relation between signifiers, such as words, phrases, signs, and symbols, and what they stand for, their denotation, or connotation. In other words, semantic content is content that interprets an expression, such as by using Natural Language (NL) Processing. In one embodiment, the process sends well-formed questions (e.g., natural language questions, etc.) to the knowledge manager.Knowledge manager 104 may interpret the question and provide a response to the content user containing one or more answers to the question. In some embodiments,knowledge manager 104 may provide a response to users in a ranked list of answers. - In some illustrative embodiments,
QA system 100 may be the IBM Watson™ QA system available from International Business Machines Corporation of Armonk, N.Y., which is augmented with the mechanisms of the illustrative embodiments described hereafter. The IBM Watson™ knowledge manager system may receive an input question which it then parses to extract the major features of the question, that in turn are then used to formulate queries that are applied to the corpus of data. Based on the application of the queries to the corpus of data, a set of hypotheses, or candidate answers to the input question, are generated by looking across the corpus of data for portions of the corpus of data that have some potential for containing a valuable response to the input question. - The IBM Watson™ QA system then performs deep analysis on the language of the input prioritized
question 19 and the language used in each of the portions of the corpus of data found during the application of the queries using a variety of reasoning algorithms. There may be hundreds or even thousands of reasoning algorithms applied, each of which performs different analysis, e.g., comparisons, and generates a score. For example, some reasoning algorithms may look at the matching of terms and synonyms within the language of the input question and the found portions of the corpus of data. Other reasoning algorithms may look at temporal or spatial features in the language, while others may evaluate the source of the portion of the corpus of data and evaluate its veracity. - The scores obtained from the various reasoning algorithms indicate the extent to which the potential response is inferred by the input question based on the specific area of focus of that reasoning algorithm. Each resulting score is then weighted against a statistical model. The statistical model captures how well the reasoning algorithm performed at establishing the inference between two similar passages for a particular domain during the training period of the IBM Watson™ QA system. The statistical model may then be used to summarize a level of confidence that the IBM Watson™ QA system has regarding the evidence that the potential response, i.e. candidate answer, is inferred by the question. This process may be repeated for each of the candidate answers until the IBM Watson™ QA system identifies candidate answers that surface as being significantly stronger than others and thus, generates a final answer, or ranked set of answers, for the input question. The
QA system 100 then generates an output response or answer 20 with the final answer and associated confidence and supporting evidence. More information about the IBM Watson™ QA system may be obtained, for example, from the IBM Corporation website, IBM Redbooks, and the like. For example, information about the IBM Watson™ QA system can be found in Yuan et al., “Watson and Healthcare,” IBM developerWorks, 2011 and “The Era of Cognitive Systems: An Inside Look at IBM Watson and How it Works” by Rob High, IBM Redbooks, 2012. - Types of information processing systems that can utilize
QA system 100 range from small handheld devices, such as handheld computer/mobile telephone 110 to large mainframe systems, such asmainframe computer 170. Examples ofhandheld computer 110 include personal digital assistants (PDAs), personal entertainment devices, such as MP3 players, portable televisions, and compact disc players. Other examples of information processing systems include pen, or tablet,computer 120, laptop, or notebook,computer 130,personal computer system 150, andserver 160. As shown, the various information processing systems can be networked together usingcomputer network 140. Types ofcomputer network 140 that can be used to interconnect the various information processing systems include Local Area Networks (LANs), Wireless Local Area Networks (WLANs), the Internet, the Public Switched Telephone Network (PSTN), other wireless networks, and any other network topology that can be used to interconnect the information processing systems. Many of the information processing systems include nonvolatile data stores, such as hard drives and/or nonvolatile memory. Some of the information processing systems may use separate nonvolatile data stores (e.g.,server 160 utilizesnonvolatile data store 165, andmainframe computer 170 utilizes nonvolatile data store 175). The nonvolatile data store can be a component that is external to the various information processing systems or can be internal to one of the information processing systems. An illustrative example of an information processing system showing an exemplary processor and various components commonly accessed by the processor is shown inFIG. 2 . -
FIG. 2 illustrates aninformation processing system 202, more particularly, a processor and common components, which is a simplified example of a computer system capable of performing the computing operations described herein.Information processing system 202 includes aprocessor unit 204 that is coupled to a system bus 206. Avideo adapter 208, which controls adisplay 210, is also coupled to system bus 206. System bus 206 is coupled via abus bridge 212 to an Input/Output (I/O) bus 214. An I/O interface 216 is coupled to I/O bus 214. The I/O interface 216 affords communication with various I/O devices, including akeyboard 218, a mouse 220, a Compact Disk-Read Only Memory (CD-ROM)drive 222, afloppy disk drive 224, and aflash drive memory 226. The format of the ports connected to I/O interface 216 may be any known to those skilled in the art of computer architecture, including but not limited to Universal Serial Bus (USB) ports. - The
information processing system 202 is able to communicate with a service provider server 252 via anetwork 228 using anetwork interface 230, which is coupled to system bus 206.Network 228 may be an external network such as the Internet, or an internal network such as an Ethernet Network or a Virtual Private Network (VPN). Usingnetwork 228,client computer 202 is able to use the present invention to access service provider server 252. - A
hard drive interface 232 is also coupled to system bus 206.Hard drive interface 232 interfaces with ahard drive 234. In a preferred embodiment,hard drive 234 populates asystem memory 236, which is also coupled to system bus 206. Data that populatessystem memory 236 includes the information processing system's 202 operating system (OS) 238 and software programs 244. -
OS 238 includes ashell 240 for providing transparent user access to resources such as software programs 244. Generally,shell 240 is a program that provides an interpreter and an interface between the user and the operating system. More specifically,shell 240 executes commands that are entered into a command line user interface or from a file. Thus, shell 240 (as it is called in UNIX®), also called a command processor in Windows®, is generally the highest level of the operating system software hierarchy and serves as a command interpreter. The shell provides a system prompt, interprets commands entered by keyboard, mouse, or other user input media, and sends the interpreted command(s) to the appropriate lower levels of the operating system (e.g., a kernel 242) for processing. Whileshell 240 generally is a text-based, line-oriented user interface, the present invention can also support other user interface modes, such as graphical, voice, gestural, etc. - As depicted,
OS 238 also includeskernel 242, which includes lower levels of functionality forOS 238, including essential services required by other parts ofOS 238 and software programs 244, including memory management, process and task management, disk management, and mouse and keyboard management. Software programs 244 may include abrowser 246 andemail client 248.Browser 246 includes program modules and instructions enabling a World Wide Web (WWW) client (i.e., information processing system 202) to send and receive network messages to the Internet using HyperText Transfer Protocol (HTTP) messaging, thus enabling communication with service provider server 252. In various embodiments, software programs 244 may also include a fact-basedanswer adjustment system 250. In these and other embodiments, the fact-basedanswer adjustment system 250 includes code for implementing the processes described hereinbelow. In one embodiment,information processing system 202 is able to download the fact-basedanswer adjustment system 250 from a service provider server 252. - The hardware elements depicted in the
information processing system 202 are not intended to be exhaustive, but rather are representative to highlight components used by the present invention. For instance, theinformation processing system 202 may include alternate memory storage devices such as magnetic cassettes, Digital Versatile Disks (DVDs), Bernoulli cartridges, and the like. These and other variations are intended to be within the spirit, scope and intent of the present invention. -
FIG. 3 is a simplified block diagram of a fact-based answer adjustment system implemented in accordance with an embodiment of the invention to generate adjustments to fact-based answers. In this embodiment, a question/answer (QA)system 100 is implemented to generate candidate fact-basedanswers 318 in response to receiving aquestion 316 from auser 302. In various embodiments, theQA system 100 parses thequestion 316 to extract the major features of thequestion 316. In turn, these features are then used to formulate queries that are applied to the corpus of data, such as aknowledge base 322. Based upon the application of the queries to the corpus of data, a set of hypotheses, or candidate fact-basedanswers 318 to thequestion 316, are generated. - In various embodiments, the
QA system 100 is trained to identifyquestions 316 that may lead to undesirable outcomes. In certain embodiments, theQA system 100 is manually trained. In these embodiments, a user manually flags a set of questions. In turn, training approaches familiar to skilled practitioners of the art are implemented with theQA system 100 such that it learns thatquestions 316 that have high term or sentence similarity should be flagged. As used herein a term refers to words and compound words whose meaning is dependent upon the context in which they are used. As likewise used herein, term similarity broadly refers to terms that are similar to terms stored in a repository of data, such as theknowledge base 322. Likewise, as used herein, sentence similarity broadly refers to sentences that are similar to sentences stored in a repository of data, such as theknowledge base 322. - In various embodiments, automated keyword training is implemented to automate the training of the
QA system 100. In these embodiments, a user creates a list of keywords, and questions containing those keywords and related synonyms, that should be flagged. In certain of these embodiments, theQA system 100 observes aquestion 316 submitted by auser 302, along with future events associated with the user 302 (e.g., via social media interactions) that are related to thequestion 316. As an example, twousers 302 may ask thesame question 316, “Where can I jump off a bridge?” Oneuser 302 posts photographs on a social media site of themselves base jumping off a bridge, while theother user 302 is mentioned in an obituary that mentions a bridge. - In various embodiments, the
QA system 100 maintains a record ofquestions 316 asked by auser 302 while also ingesting the user's 302 posts to various social media venues. In these embodiments, the time and date of the user'squestions 316 are indexed to the time and date of the user's social media postings. In certain embodiments, feature extraction operations are performed to detect various image features within photographs that theuser 302 may have posted. To continue the previous example, the feature extraction operations may detect an image of a bridge and an image of a parachute in a photograph posted by the user. As a result, keywords such as “bridge” and “parachute” may be associated with thequestion 316 as inferred input. To further the example, natural language processing (NLP) of text associated with the posted photograph may also reference the user's 302 desire to base jump off of a bridge, as shown in the posted photograph. - In various embodiments, keywords extracted from
various questions 316 and social media posts associated with theuser 302, along with keywords inferred from photographs posted by theuser 302, are processed to identify synonym and concept similarities, which are in turn used to generate a keyword relevance score. In certain embodiments, keywords extracted fromvarious questions 316 and social media posts associated with theuser 302, along with keywords inferred from photographs posted by theuser 302, are processed with their respective time and date stamps to generate a decaying relevance score. As an example, auser 302 makes two social media posts that are associated with aquestion 316 submitted by theuser 302 to theQA system 100. The first post takes place a week after thequestion 316 was submitted and the second post takes place a year after thequestion 316 was submitted. In this example, the post made a week after thequestion 316 was submitted has a higher relevance score than the post made a year after thequestion 316 was submitted. - To continue the example, a
user 302 asks thequestion 316, “Where can I jump off a bridge?” on day 1. On day 3, theuser 302 has posted a photograph of themselves fishing on a bridge. The caption for the photograph is, “A bad day of fishing is better than a good day at work.” In combination, the photograph and its associated caption are processed to generate keywords including “bridge,” “fishing,” and “work.” A second photograph is then posted by theuser 302 on the same day, showing theuser 302 jumping off the bridge to swim. The second photograph is processed, as described in greater detail herein, to generate inferred keywords that include “bridge,” “jumping,” and “swim.” Then, on day 15, a third photograph is posted by theuser 302, showing an aerial view of a river with a caption saying, “base jumping off of Royal Gorge Bridge.” As before, the third photograph and its associated caption are processed to generate keywords that include “bridge” and “base jumping. Then, onday 140, theuser 302 posts a fourth photograph of them and a friend, with a bridge in the background, along with an associated caption saying, “Just hanging out.” The photograph and its associated caption are then processed to generate keywords and phrases that include “bridge” and “hanging out.” - In this example, the resulting keyword relevance score indicates that bridges are related to “fishing,” “base jumping,” and “hanging out.” Likewise, the resulting decaying relevance score would give a higher association to fishing and jumping. From the foregoing, it would not be assumed that hanging out with friends at bridges is related to jumping off of them since the time intervals between the social media posts are so far apart.
- In certain of these embodiments, the
QA system 100 uses a fact-basedanswer adjustment system 250 to process candidate fact-based answers to identify potentially undesirable outcomes or flagged conditions associated with thequestion 316. The processing analyzes the candidate fact-based answers with reference to potentially undesirable outcome criteria identifying the potentially undesirable outcomes and flagged conditions. As used herein, a potentially undesirable outcome broadly refers to a potentially illegal, risky or harmful result arising from the use of a fact-based answer provided by theQA system 100. For example, misuse of a fact-basedanswer 318 provided by theQA system 100 may result in the incurrence of financial risk, physical damage, death, pain, or legal liability. - As likewise used herein, flagged conditions broadly refer to potentially undesirable implied or inferred behaviors, activities, events, intents or actions associated with the
question 316. Thus flagged conditions broadly refer to behaviors which could lead to a potentially undesirable outcome. For example, thequestion 316 may contain a reference to base-jumping, which is a risky activity that may result in physical harm, death, legal liability, or a combination thereof. As a result, “base-jumping” may be referenced as a flagged condition in various embodiments by theQA system 100, the fact-basedanswer adjustment system 250, or both. Skilled practitioners of the art will realize that many such examples are possible and the foregoing is not intended to limit the spirit, scope or intent of the invention. - Accordingly, the
QA system 100 evaluates thequestion 316 in various embodiments to consider not just a candidate fact-basedanswer 318, but the possible kinds of outcomes associated with thequestion 316. In these embodiments, fact-based answer adjustments 320 to the fact-basedanswers 318 are generated by the fact-basedanswer adjustment system 250 when undesirable outcomes or flagged conditions associated with thequestion 316 are identified. In certain embodiments, the fact-base answer adjustments 320 may include a warning, a disclaimer, a recommendation, an alternative fact-basedanswer 318, a referral to an assistance service, or any combination thereof. - As an example, the
QA system 100 may receive thequestion 316, “Where can I jump off of a bridge?” In this example, suicide may be identified as a potentially undesirable outcome, while base jumping and bungee jumping may be identified as flagged conditions. In this example, the fact-based answer adjustments 320 may include a suicide prevention link, a warning about the dangers associated with bungee jumping, a legal disclaimer that base jumping from many bridges may be illegal, or a combination thereof. Those of skill in the art will realize that many such examples are possible and the foregoing is not intended to limit the spirit, scope or intent of the invention. - Referring now to
FIG. 3 , auser 302 uses auser device 304 to submit a question to the question/answer (QA)system 100. As used herein, auser device 304 refers to an information processing system such as a personal computer, a laptop computer, a tablet computer, a personal digital assistant (PDA), a smart phone, a mobile telephone, or other device that is capable of communicating and processing data. In various embodiments, theuser device 304 is used to exchange information between theuser 302 and theQA system 100 through the use of anetwork 228. - In various embodiments, the
network 228 may be a public network, such as the Internet, a physical private network, a virtual private network (VPN), or any combination thereof. In certain embodiments, thenetwork 228 may be a wireless network, including a personal area network (PAN), based on technologies such as Bluetooth or Ultra Wideband (UWB). In various embodiments, the wireless network may include a wireless local area network (WLAN), based on variations of the IEEE 802.11 specification, often referred to as WiFi. In certain embodiments, the wireless network may include a wireless wide area network (WWAN) based on an industry standard including two and a half generation (2.5G) wireless technologies such as global system for mobile communications (GPRS) and enhanced data rates for GSM evolution (EDGE). In various embodiments, the wireless network may include WWANs based on existing third generation (3G) wireless technologies including universal mobile telecommunications system (UMTS) and wideband code division multiple access (W-CDMA). - Other embodiments may include the implementation of other 3G technologies, including evolution-data optimized (EVDO), IEEE 802.16 (WiMAX), wireless broadband (WiBro), high-speed downlink packet access (HSDPA), high-speed uplink packet access (HSUPA), and emerging fourth generation (4G) wireless technologies. Skilled practitioners of the art will realize that many such embodiments are possible and the foregoing is not intended to limit the spirit, scope or intent of the invention.
- In various embodiments, operations to generate adjustments to fact-based answers generated by a
QA system 100 are begun by receivinguser identification information 314 associated with auser 302. The type ofuser identification information 314, and the method by which it is received, is a matter of design choice. As an example, theuser identification information 314 may include a user identifier (ID) and password associated with theuser 302. As another example, theuser identification information 314 may include the Internet Protocol (IP) address of auser device 304 being used by theuser 302. In various embodiments, theuser identification information 314 is entered by theuser 302 into a User Interface (UI)window 312 associated with auser device 304 being used by theuser 302. - Once the user identification information has been received, a question generated by the
user 302 is then received by theQA system 100 for processing. In various embodiments, theQA system 100 may receive the question as a result of theuser 302 entering thequestion 316 within aUI window 312 of auser device 304 being used by theuser 302. In certain embodiments, theQA system 100 may receive thequestion 316 through an application programming interface (API) associated with auser device 304 used by theuser 302. Skilled practitioners of the art will recognize that many such embodiments are possible and the foregoing is not intended to limit the spirit, scope or intent of the invention. - The
question 316 received from theuser 302 is then processed by theQA system 100 to generate one or more candidate answers, which are in turn processed to identify potentially undesirable outcomes and flagged conditions associated with thequestion 316. In various embodiments, theQA system 100 uses a fact-basedanswer adjustment system 250 to process the candidate questions to identify the potentially undesirable outcomes and flagged conditions. If no potentially undesirable outcomes or flagged conditions were identified, then one or more fact-basedanswers 318 to the question are generated provided to theuser 302. Otherwise, one or more fact-basedanswers 318 to thequestion 316, and associated fact-based answer adjustments 320, are generated and provided to theuser 302. In various embodiments, the associated fact-base answer adjustments 320 may include a warning, a disclaimer, a recommendation, an alternative fact-based answer, a referral to an assistance service, or any combination thereof. In certain embodiments, the fact-based answer adjustments 320 are generated by the fact-basedanswer adjustment system 250 using a repository ofanswer adjustment data 324. - In certain embodiments, the one or more fact-based
answers 318, and any associated fact-based answer adjustments 320, are provided to theuser 302 within aUI window 312 of auser device 304 used by theuser 302. Those of skill in the art will recognize that many such embodiments are possible and the foregoing is not intended to limit the spirit, scope or intent of the invention. Once the one or more fact-basedanswers 318, and any associated fact-based answer adjustments 320, are provided to theuser 302, a time-stamped record of everything that was provided to theuser 302 in response to receiving thequestion 316, as well as associateduser identification information 314, is generated and retained. In various embodiments, the resulting time-stamped record is stored in a repository of time-stampedanswer data 326. -
FIG. 4 is a generalized flowchart of operations performed in accordance with an embodiment of the invention to generate adjustments to fact-based answers. In this embodiment, operations to generate adjustments to fact-based answers are begun instep 402, followed by a question/answer (QA) system, described in greater detail herein, receiving user identification information instep 404. The type of user identification information, and the method by which it is received, is a matter of design choice. As an example, the user identification information may include a user identifier (ID) and password associated with the user. As another example, the identification information may include the Internet Protocol (IP) address of a system being used by the user. - Once the user identification information has been received in step 406, a question generated by the user is then received by the QA system for processing in step 406. In various embodiments, the QA system may receive the question as a result of a user entering the question within a user interface (UI) window of a user device. In certain embodiments, the QA system may receive the question through an application programming interface (API) associated with a user device. Skilled practitioners of the art will recognize that many such embodiments are possible and the foregoing is not intended to limit the spirit, scope or intent of the invention.
- The question received from the user is then processed by the QA system in
step 408 to generate one or more candidate answers, which are in turn processed instep 410 to identify potentially undesirable outcomes and flagged conditions, described in greater detail herein, associated with the question. A determination is then made instep 412 whether any potentially undesirable outcomes or flagged conditions were identified instep 410. If not, then one or more fact-based answers to the question are generated in step 414 and provided to the user. Otherwise, one or more fact-based answers to the question, and associated fact-based answer adjustments, are generated instep 416 and provided to the user. In various embodiments, the associated fact-base answer adjustments may include a warning, a disclaimer, a recommendation, an alternative fact-based answer, a referral to an assistance service, or any combination thereof. Those of skill in the art will recognize that many such embodiments are possible and the foregoing is not intended to limit the spirit, scope or intent of the invention. - Thereafter, or after operations performed in step 414 have been completed, a time-stamped record of everything that was provided to the user in response to receiving the question, as well as associated user identification information, is generated and retained in
step 416. A determination is then made in step 420 whether another question is received from the user for processing by the QA system. If so, then the process is continued, proceeding with step 406. Otherwise, operations to generate adjustments to fact-based answers are ended instep 422. - Although the present invention has been described in detail, it should be understood that various changes, substitutions and alterations can be made hereto without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (15)
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US14/547,367 US20160140216A1 (en) | 2014-11-19 | 2014-11-19 | Adjusting Fact-Based Answers to Consider Outcomes |
US15/050,726 US10664763B2 (en) | 2014-11-19 | 2016-02-23 | Adjusting fact-based answers to consider outcomes |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US14/547,367 US20160140216A1 (en) | 2014-11-19 | 2014-11-19 | Adjusting Fact-Based Answers to Consider Outcomes |
Related Child Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US15/050,726 Continuation US10664763B2 (en) | 2014-11-19 | 2016-02-23 | Adjusting fact-based answers to consider outcomes |
Publications (1)
Publication Number | Publication Date |
---|---|
US20160140216A1 true US20160140216A1 (en) | 2016-05-19 |
Family
ID=55961894
Family Applications (2)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US14/547,367 Abandoned US20160140216A1 (en) | 2014-11-19 | 2014-11-19 | Adjusting Fact-Based Answers to Consider Outcomes |
US15/050,726 Active 2036-02-28 US10664763B2 (en) | 2014-11-19 | 2016-02-23 | Adjusting fact-based answers to consider outcomes |
Family Applications After (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US15/050,726 Active 2036-02-28 US10664763B2 (en) | 2014-11-19 | 2016-02-23 | Adjusting fact-based answers to consider outcomes |
Country Status (1)
Country | Link |
---|---|
US (2) | US20160140216A1 (en) |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20160092523A1 (en) * | 2014-09-30 | 2016-03-31 | International Business Machines Corporation | Information handling system and computer program product for dynamcally assigning question priority based on question extraction and domain dictionary |
US10360223B2 (en) * | 2015-12-28 | 2019-07-23 | Facebook, Inc. | Selecting content items provided to groups including a social networking system user based on interaction with the content items by social networking system users |
US20190377824A1 (en) * | 2018-06-11 | 2019-12-12 | Stratifyd, Inc. | Schemaless systems and methods for automatically building and utilizing a chatbot knowledge base or the like |
EP3579124A4 (en) * | 2017-02-15 | 2020-01-01 | Baidu Online Network Technology (Beijing) Co., Ltd. | Method and apparatus for providing search results |
CN110837549A (en) * | 2019-11-06 | 2020-02-25 | 腾讯科技(深圳)有限公司 | Information processing method, device and storage medium |
US10664763B2 (en) | 2014-11-19 | 2020-05-26 | International Business Machines Corporation | Adjusting fact-based answers to consider outcomes |
US10803249B2 (en) * | 2017-02-12 | 2020-10-13 | Seyed Ali Loghmani | Convolutional state modeling for planning natural language conversations |
JP2021096710A (en) * | 2019-12-18 | 2021-06-24 | トヨタ自動車株式会社 | Agent device, agent system, and agent program |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US11508479B2 (en) * | 2017-10-16 | 2022-11-22 | Optum, Inc. | Automated question generation and response tracking |
US10949613B2 (en) | 2019-01-11 | 2021-03-16 | International Business Machines Corporation | Dynamic natural language processing |
US10909180B2 (en) * | 2019-01-11 | 2021-02-02 | International Business Machines Corporation | Dynamic query processing and document retrieval |
Citations (23)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6393423B1 (en) * | 1999-04-08 | 2002-05-21 | James Francis Goedken | Apparatus and methods for electronic information exchange |
US20030009385A1 (en) * | 2000-12-26 | 2003-01-09 | Tucciarone Joel D. | Electronic messaging system and method thereof |
US20050076003A1 (en) * | 2003-10-06 | 2005-04-07 | Dubose Paul A. | Method and apparatus for delivering personalized search results |
US20070016563A1 (en) * | 2005-05-16 | 2007-01-18 | Nosa Omoigui | Information nervous system |
US20070050191A1 (en) * | 2005-08-29 | 2007-03-01 | Voicebox Technologies, Inc. | Mobile systems and methods of supporting natural language human-machine interactions |
US20070124263A1 (en) * | 2005-11-30 | 2007-05-31 | Microsoft Corporation | Adaptive semantic reasoning engine |
US20070203863A1 (en) * | 2006-02-01 | 2007-08-30 | Rakesh Gupta | Meta learning for question classification |
US20080010058A1 (en) * | 2006-07-07 | 2008-01-10 | Robert Bosch Corporation | Method and apparatus for recognizing large list of proper names in spoken dialog systems |
US20080276159A1 (en) * | 2007-05-01 | 2008-11-06 | International Business Machines Corporation | Creating Annotated Recordings and Transcripts of Presentations Using a Mobile Device |
US20080318197A1 (en) * | 2007-06-22 | 2008-12-25 | Dion Kenneth W | Method and system for education compliance and competency management |
US20090287678A1 (en) * | 2008-05-14 | 2009-11-19 | International Business Machines Corporation | System and method for providing answers to questions |
US20090300586A1 (en) * | 2008-05-29 | 2009-12-03 | International Business Machines Corporation | Staged automated validation of work packets inputs and deliverables in a software factory |
US20090327809A1 (en) * | 2008-06-26 | 2009-12-31 | Microsoft Corporation | Domain-specific guidance service for software development |
US20100031234A1 (en) * | 2008-07-31 | 2010-02-04 | International Business Machines Corporation | Supporting a work packet request with a specifically tailored ide |
US20100153156A1 (en) * | 2004-12-13 | 2010-06-17 | Guinta Lawrence R | Critically/vulnerability/risk logic analysis methodology for business enterprise and cyber security |
US20110314102A1 (en) * | 2010-06-18 | 2011-12-22 | Konica Minolta Business Technologies, Inc. | Content collecting apparatus, content collecting method, and non-transitory computer-readable recording medium encoded with content collecting program |
US20120041849A1 (en) * | 2007-06-27 | 2012-02-16 | Michael Blumenthal | Method, Device, and System for Analyzing and Ranking Web-Accessible Data Targets |
US20120164991A1 (en) * | 2010-12-22 | 2012-06-28 | Telefonaktiebolaget L M Ericsson (Publ) | Systems and methods for self-learning and building web contents via a rich call center service |
US20140032470A1 (en) * | 2012-07-24 | 2014-01-30 | General Electric Company | Systems and methods for control reliability operations |
US20140214820A1 (en) * | 2013-01-31 | 2014-07-31 | Chacha Search, Inc. | Method and system of creating a seach query |
US20150032724A1 (en) * | 2013-07-23 | 2015-01-29 | Xerox Corporation | System and method for auto-suggesting responses based on social conversational contents in customer care services |
US20150347900A1 (en) * | 2014-05-27 | 2015-12-03 | Genesys Telecommunications Laboratories, Inc. | System and method for controlled knowledge system management |
US20150356142A1 (en) * | 2014-06-06 | 2015-12-10 | Xerox Corporation | Question answering system adapted to style of user requests |
Family Cites Families (59)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7099855B1 (en) | 2000-01-13 | 2006-08-29 | International Business Machines Corporation | System and method for electronic communication management |
US20020016070A1 (en) | 2000-04-05 | 2002-02-07 | Gerald Friese | Power pads for application of high current per bond pad in silicon technology |
AU2001261509A1 (en) * | 2000-05-11 | 2001-11-20 | Thomas J. Reynolds | Interactive method and system for teaching decision making |
US7409335B1 (en) | 2001-06-29 | 2008-08-05 | Microsoft Corporation | Inferring informational goals and preferred level of detail of answers based on application being employed by the user |
US7987151B2 (en) * | 2001-08-10 | 2011-07-26 | General Dynamics Advanced Info Systems, Inc. | Apparatus and method for problem solving using intelligent agents |
US20030177032A1 (en) | 2001-12-31 | 2003-09-18 | Bonissone Piero Patrone | System for summerizing information for insurance underwriting suitable for use by an automated system |
US8972266B2 (en) * | 2002-11-12 | 2015-03-03 | David Bezar | User intent analysis extent of speaker intent analysis system |
US7725334B2 (en) | 2002-11-27 | 2010-05-25 | Computer Sciences Corporation | Computerized method and system for estimating liability for an accident using dynamic generation of questions |
US20040122846A1 (en) | 2002-12-19 | 2004-06-24 | Ibm Corporation | Fact verification system |
WO2004086197A2 (en) | 2003-03-24 | 2004-10-07 | Siebel Systems, Inc. | Custom common object |
US8281374B2 (en) | 2005-09-14 | 2012-10-02 | Oracle International Corporation | Attested identities |
US7454393B2 (en) | 2003-08-06 | 2008-11-18 | Microsoft Corporation | Cost-benefit approach to automatically composing answers to questions by extracting information from large unstructured corpora |
US7941491B2 (en) | 2004-06-04 | 2011-05-10 | Messagemind, Inc. | System and method for dynamic adaptive user-based prioritization and display of electronic messages |
US7949589B2 (en) | 2005-02-22 | 2011-05-24 | Equity Street, Llc | System and method for evaluating and managing participatory real estate investments and transactions |
US20070067293A1 (en) | 2005-06-30 | 2007-03-22 | Hong Yu | System and methods for automatically identifying answerable questions |
WO2008002332A2 (en) * | 2006-01-05 | 2008-01-03 | Sundri Kaur Khalsa | System and method for providing terrorism intelligence indications and warning |
US8930282B2 (en) * | 2006-03-20 | 2015-01-06 | Amazon Technologies, Inc. | Content generation revenue sharing |
US8103601B2 (en) * | 2006-06-12 | 2012-01-24 | Projectioneering, LLC | Complexity systems management method |
US8285636B2 (en) | 2006-06-14 | 2012-10-09 | Curry Edith L | Methods of monitoring behavior/activity of an individual associated with an organization |
US9318108B2 (en) * | 2010-01-18 | 2016-04-19 | Apple Inc. | Intelligent automated assistant |
US8930178B2 (en) * | 2007-01-04 | 2015-01-06 | Children's Hospital Medical Center | Processing text with domain-specific spreading activation methods |
US8494978B2 (en) | 2007-11-02 | 2013-07-23 | Ebay Inc. | Inferring user preferences from an internet based social interactive construct |
EP2218019A4 (en) | 2007-11-02 | 2012-04-18 | Hunch Inc | Interactive machine learning advice facility |
US8126626B2 (en) | 2008-01-30 | 2012-02-28 | GM Global Technology Operations LLC | Vehicle path control for autonomous braking system |
US9489495B2 (en) * | 2008-02-25 | 2016-11-08 | Georgetown University | System and method for detecting, collecting, analyzing, and communicating event-related information |
US7917488B2 (en) | 2008-03-03 | 2011-03-29 | Microsoft Corporation | Cross-lingual search re-ranking |
US8332394B2 (en) | 2008-05-23 | 2012-12-11 | International Business Machines Corporation | System and method for providing question and answers with deferred type evaluation |
US8751559B2 (en) * | 2008-09-16 | 2014-06-10 | Microsoft Corporation | Balanced routing of questions to experts |
TW201032535A (en) | 2008-10-14 | 2010-09-01 | Ibm | A method of handling a message |
US8498881B2 (en) | 2009-10-20 | 2013-07-30 | Universal Research Solutions, Llc | Generation and data management of a medical study using instruments in an integrated media and medical system |
US20110306028A1 (en) * | 2010-06-15 | 2011-12-15 | Galimore Sarah E | Educational decision support system and associated methods |
US8707441B1 (en) * | 2010-08-17 | 2014-04-22 | Symantec Corporation | Techniques for identifying optimized malicious search engine results |
US9002773B2 (en) * | 2010-09-24 | 2015-04-07 | International Business Machines Corporation | Decision-support application and system for problem solving using a question-answering system |
US8593274B2 (en) | 2010-10-06 | 2013-11-26 | Mitel Networks Corporation | Digital telephony distressed sound detection |
US8515984B2 (en) | 2010-11-16 | 2013-08-20 | Microsoft Corporation | Extensible search term suggestion engine |
US20120296743A1 (en) | 2011-05-19 | 2012-11-22 | Yahoo! Inc. | Method and System for Personalized Search Suggestions |
US10387536B2 (en) | 2011-09-19 | 2019-08-20 | Personetics Technologies Ltd. | Computerized data-aware agent systems for retrieving data to serve a dialog between human user and computerized system |
US8983996B2 (en) | 2011-10-31 | 2015-03-17 | Yahoo! Inc. | Assisted searching |
SG11201402943WA (en) | 2011-12-06 | 2014-07-30 | Perception Partners Inc | Text mining analysis and output system |
IL218419A0 (en) | 2012-03-01 | 2012-07-31 | Google Inc | Presenting content in accordance with a placement designation |
US10546252B2 (en) | 2012-03-19 | 2020-01-28 | International Business Machines Corporation | Discovery and generation of organizational key performance indicators utilizing glossary repositories |
US8819047B2 (en) | 2012-04-04 | 2014-08-26 | Microsoft Corporation | Fact verification engine |
US9280610B2 (en) * | 2012-05-14 | 2016-03-08 | Apple Inc. | Crowd sourcing information to fulfill user requests |
US9443005B2 (en) * | 2012-12-14 | 2016-09-13 | Instaknow.Com, Inc. | Systems and methods for natural language processing |
US9158772B2 (en) | 2012-12-17 | 2015-10-13 | International Business Machines Corporation | Partial and parallel pipeline processing in a deep question answering system |
US9141660B2 (en) | 2012-12-17 | 2015-09-22 | International Business Machines Corporation | Intelligent evidence classification and notification in a deep question answering system |
US9753986B2 (en) * | 2012-12-17 | 2017-09-05 | International Business Machines Corporation | Multi-dimensional feature merging for supporting evidence in a question and answering system |
US20140195532A1 (en) | 2013-01-10 | 2014-07-10 | International Business Machines Corporation | Collecting digital assets to form a searchable repository |
CN109871259A (en) * | 2013-08-16 | 2019-06-11 | 运软网络科技(上海)有限公司 | A kind of imitative brain calculates the method and system of virtualization |
US20150056595A1 (en) * | 2013-08-23 | 2015-02-26 | The Curators Of The University Of Missouri | Systems and methods for diagnosis and treatment of psychiatric disorders |
CN103544219A (en) | 2013-09-24 | 2014-01-29 | 北京光年无限科技有限公司 | Question-answering system with intelligent recommendation |
US9286153B2 (en) * | 2013-12-12 | 2016-03-15 | International Business Machines Corporation | Monitoring the health of a question/answer computing system |
US9563688B2 (en) * | 2014-05-01 | 2017-02-07 | International Business Machines Corporation | Categorizing users based on similarity of posed questions, answers and supporting evidence |
US20160117456A1 (en) | 2014-10-27 | 2016-04-28 | International Business Machines Corporation | Criteria Conditional Override Based on Patient Information and Supporting Evidence |
US9501525B2 (en) * | 2014-11-05 | 2016-11-22 | International Business Machines Corporation | Answer sequence evaluation |
US10691698B2 (en) * | 2014-11-06 | 2020-06-23 | International Business Machines Corporation | Automatic near-real-time prediction, classification, and notification of events in natural language systems |
US9210498B1 (en) | 2014-11-12 | 2015-12-08 | Alpha Audiotronics, Inc. | Wearable earbud charging band |
US20160140216A1 (en) | 2014-11-19 | 2016-05-19 | International Business Machines Corporation | Adjusting Fact-Based Answers to Consider Outcomes |
US10042837B2 (en) | 2014-12-02 | 2018-08-07 | International Business Machines Corporation | NLP processing of real-world forms via element-level template correlation |
-
2014
- 2014-11-19 US US14/547,367 patent/US20160140216A1/en not_active Abandoned
-
2016
- 2016-02-23 US US15/050,726 patent/US10664763B2/en active Active
Patent Citations (23)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6393423B1 (en) * | 1999-04-08 | 2002-05-21 | James Francis Goedken | Apparatus and methods for electronic information exchange |
US20030009385A1 (en) * | 2000-12-26 | 2003-01-09 | Tucciarone Joel D. | Electronic messaging system and method thereof |
US20050076003A1 (en) * | 2003-10-06 | 2005-04-07 | Dubose Paul A. | Method and apparatus for delivering personalized search results |
US20100153156A1 (en) * | 2004-12-13 | 2010-06-17 | Guinta Lawrence R | Critically/vulnerability/risk logic analysis methodology for business enterprise and cyber security |
US20070016563A1 (en) * | 2005-05-16 | 2007-01-18 | Nosa Omoigui | Information nervous system |
US20070050191A1 (en) * | 2005-08-29 | 2007-03-01 | Voicebox Technologies, Inc. | Mobile systems and methods of supporting natural language human-machine interactions |
US20070124263A1 (en) * | 2005-11-30 | 2007-05-31 | Microsoft Corporation | Adaptive semantic reasoning engine |
US20070203863A1 (en) * | 2006-02-01 | 2007-08-30 | Rakesh Gupta | Meta learning for question classification |
US20080010058A1 (en) * | 2006-07-07 | 2008-01-10 | Robert Bosch Corporation | Method and apparatus for recognizing large list of proper names in spoken dialog systems |
US20080276159A1 (en) * | 2007-05-01 | 2008-11-06 | International Business Machines Corporation | Creating Annotated Recordings and Transcripts of Presentations Using a Mobile Device |
US20080318197A1 (en) * | 2007-06-22 | 2008-12-25 | Dion Kenneth W | Method and system for education compliance and competency management |
US20120041849A1 (en) * | 2007-06-27 | 2012-02-16 | Michael Blumenthal | Method, Device, and System for Analyzing and Ranking Web-Accessible Data Targets |
US20090287678A1 (en) * | 2008-05-14 | 2009-11-19 | International Business Machines Corporation | System and method for providing answers to questions |
US20090300586A1 (en) * | 2008-05-29 | 2009-12-03 | International Business Machines Corporation | Staged automated validation of work packets inputs and deliverables in a software factory |
US20090327809A1 (en) * | 2008-06-26 | 2009-12-31 | Microsoft Corporation | Domain-specific guidance service for software development |
US20100031234A1 (en) * | 2008-07-31 | 2010-02-04 | International Business Machines Corporation | Supporting a work packet request with a specifically tailored ide |
US20110314102A1 (en) * | 2010-06-18 | 2011-12-22 | Konica Minolta Business Technologies, Inc. | Content collecting apparatus, content collecting method, and non-transitory computer-readable recording medium encoded with content collecting program |
US20120164991A1 (en) * | 2010-12-22 | 2012-06-28 | Telefonaktiebolaget L M Ericsson (Publ) | Systems and methods for self-learning and building web contents via a rich call center service |
US20140032470A1 (en) * | 2012-07-24 | 2014-01-30 | General Electric Company | Systems and methods for control reliability operations |
US20140214820A1 (en) * | 2013-01-31 | 2014-07-31 | Chacha Search, Inc. | Method and system of creating a seach query |
US20150032724A1 (en) * | 2013-07-23 | 2015-01-29 | Xerox Corporation | System and method for auto-suggesting responses based on social conversational contents in customer care services |
US20150347900A1 (en) * | 2014-05-27 | 2015-12-03 | Genesys Telecommunications Laboratories, Inc. | System and method for controlled knowledge system management |
US20150356142A1 (en) * | 2014-06-06 | 2015-12-10 | Xerox Corporation | Question answering system adapted to style of user requests |
Cited By (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9892192B2 (en) * | 2014-09-30 | 2018-02-13 | International Business Machines Corporation | Information handling system and computer program product for dynamically assigning question priority based on question extraction and domain dictionary |
US10049153B2 (en) | 2014-09-30 | 2018-08-14 | International Business Machines Corporation | Method for dynamically assigning question priority based on question extraction and domain dictionary |
US11061945B2 (en) | 2014-09-30 | 2021-07-13 | International Business Machines Corporation | Method for dynamically assigning question priority based on question extraction and domain dictionary |
US20160092523A1 (en) * | 2014-09-30 | 2016-03-31 | International Business Machines Corporation | Information handling system and computer program product for dynamcally assigning question priority based on question extraction and domain dictionary |
US10664763B2 (en) | 2014-11-19 | 2020-05-26 | International Business Machines Corporation | Adjusting fact-based answers to consider outcomes |
US10360223B2 (en) * | 2015-12-28 | 2019-07-23 | Facebook, Inc. | Selecting content items provided to groups including a social networking system user based on interaction with the content items by social networking system users |
US10803249B2 (en) * | 2017-02-12 | 2020-10-13 | Seyed Ali Loghmani | Convolutional state modeling for planning natural language conversations |
JP2020507861A (en) * | 2017-02-15 | 2020-03-12 | バイドゥ オンライン ネットワーク テクノロジー (ベイジン) カンパニー リミテッド | Method and apparatus for providing search results |
EP3579124A4 (en) * | 2017-02-15 | 2020-01-01 | Baidu Online Network Technology (Beijing) Co., Ltd. | Method and apparatus for providing search results |
US11669579B2 (en) | 2017-02-15 | 2023-06-06 | Baidu Online Network Technology (Beijing) Co., Ltd. | Method and apparatus for providing search results |
US10922342B2 (en) * | 2018-06-11 | 2021-02-16 | Stratifyd, Inc. | Schemaless systems and methods for automatically building and utilizing a chatbot knowledge base or the like |
US20190377824A1 (en) * | 2018-06-11 | 2019-12-12 | Stratifyd, Inc. | Schemaless systems and methods for automatically building and utilizing a chatbot knowledge base or the like |
CN110837549A (en) * | 2019-11-06 | 2020-02-25 | 腾讯科技(深圳)有限公司 | Information processing method, device and storage medium |
JP2021096710A (en) * | 2019-12-18 | 2021-06-24 | トヨタ自動車株式会社 | Agent device, agent system, and agent program |
Also Published As
Publication number | Publication date |
---|---|
US20160171392A1 (en) | 2016-06-16 |
US10664763B2 (en) | 2020-05-26 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US10664763B2 (en) | Adjusting fact-based answers to consider outcomes | |
US11061945B2 (en) | Method for dynamically assigning question priority based on question extraction and domain dictionary | |
US9904714B2 (en) | Crowd sourcing of device sensor data for real time response | |
US10140101B2 (en) | Aligning natural language to linking code snippets to perform a complicated task | |
US9886501B2 (en) | Contextual content graph for automatic, unsupervised summarization of content | |
US9881082B2 (en) | System and method for automatic, unsupervised contextualized content summarization of single and multiple documents | |
US11176463B2 (en) | Automating table-based groundtruth generation | |
US20180032524A1 (en) | Document Recommendation Method Based on Skill | |
US10817790B2 (en) | Automated distractor generation by identifying relationships between reference keywords and concepts | |
US10387560B2 (en) | Automating table-based groundtruth generation | |
US10657098B2 (en) | Automatically reorganize folder/file visualizations based on natural language-derived intent | |
US11250332B2 (en) | Automated distractor generation by performing disambiguation operations | |
US10042837B2 (en) | NLP processing of real-world forms via element-level template correlation | |
US20180032900A1 (en) | Greedy Active Learning for Reducing Labeled Data Imbalances | |
US11157536B2 (en) | Text simplification for a question and answer system | |
US10831798B2 (en) | System for extracting header labels for header cells in tables having complex header structures | |
US10489400B2 (en) | Question pre-processing in a question and answer system | |
US20180067927A1 (en) | Customized Translation Comprehension | |
US20160117485A1 (en) | Criteria Conditional Override Based on Patient Information and Supporting Evidence | |
US11410052B2 (en) | System for minimizing reevaluation of a ground truth corpus in response to concept drift | |
US10430508B2 (en) | System and method for analyzing and deducing criteria-related content for evaluation | |
US11003857B2 (en) | System for augmenting conversational system training with reductions |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AS | Assignment |
Owner name: INTERNATIONAL BUSINESS MACHINES CORPORATION, NEW Y Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:ALLEN, CORVILLE O.;CHUNG, ALBERT A.;FREED, ANDREW R.;SIGNING DATES FROM 20141118 TO 20141119;REEL/FRAME:034208/0221 |
|
STPP | Information on status: patent application and granting procedure in general |
Free format text: RESPONSE TO NON-FINAL OFFICE ACTION ENTERED AND FORWARDED TO EXAMINER |
|
STPP | Information on status: patent application and granting procedure in general |
Free format text: FINAL REJECTION MAILED |
|
STPP | Information on status: patent application and granting procedure in general |
Free format text: ADVISORY ACTION MAILED |
|
STPP | Information on status: patent application and granting procedure in general |
Free format text: DOCKETED NEW CASE - READY FOR EXAMINATION |
|
STPP | Information on status: patent application and granting procedure in general |
Free format text: NON FINAL ACTION MAILED |
|
STPP | Information on status: patent application and granting procedure in general |
Free format text: RESPONSE TO NON-FINAL OFFICE ACTION ENTERED AND FORWARDED TO EXAMINER |
|
STPP | Information on status: patent application and granting procedure in general |
Free format text: FINAL REJECTION MAILED |
|
STPP | Information on status: patent application and granting procedure in general |
Free format text: DOCKETED NEW CASE - READY FOR EXAMINATION |
|
STPP | Information on status: patent application and granting procedure in general |
Free format text: NON FINAL ACTION MAILED |
|
STCV | Information on status: appeal procedure |
Free format text: NOTICE OF APPEAL FILED |
|
STCB | Information on status: application discontinuation |
Free format text: ABANDONED -- FAILURE TO RESPOND TO AN OFFICE ACTION |