WO2022201541A1 - 検索装置、検索方法、及びコンピュータ読み取り可能な記録媒体 - Google Patents
検索装置、検索方法、及びコンピュータ読み取り可能な記録媒体 Download PDFInfo
- Publication number
- WO2022201541A1 WO2022201541A1 PCT/JP2021/013097 JP2021013097W WO2022201541A1 WO 2022201541 A1 WO2022201541 A1 WO 2022201541A1 JP 2021013097 W JP2021013097 W JP 2021013097W WO 2022201541 A1 WO2022201541 A1 WO 2022201541A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- search
- information
- attribute information
- score
- search target
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims description 27
- 238000004364 calculation method Methods 0.000 claims abstract description 40
- 238000000605 extraction Methods 0.000 abstract description 30
- 239000000284 extract Substances 0.000 abstract description 13
- 230000006870 function Effects 0.000 description 75
- 239000011521 glass Substances 0.000 description 23
- 238000010586 diagram Methods 0.000 description 16
- 235000014653 Carica parviflora Nutrition 0.000 description 11
- 241000242757 Anthozoa Species 0.000 description 6
- 241000243321 Cnidaria Species 0.000 description 5
- 238000004891 communication Methods 0.000 description 4
- 238000002474 experimental method Methods 0.000 description 4
- 238000004088 simulation Methods 0.000 description 4
- 230000005540 biological transmission Effects 0.000 description 3
- 241000972773 Aulopiformes Species 0.000 description 2
- 230000003190 augmentative effect Effects 0.000 description 2
- 238000005401 electroluminescence Methods 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 235000019515 salmon Nutrition 0.000 description 2
- 239000004065 semiconductor Substances 0.000 description 2
- 241000219094 Vitaceae Species 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 239000011449 brick Substances 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 235000021021 grapes Nutrition 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000010422 painting Methods 0.000 description 1
- 238000007639 printing Methods 0.000 description 1
Images
Classifications
-
- 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
- G06F16/334—Query execution
-
- 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
- G06F16/338—Presentation of query results
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/53—Querying
- G06F16/532—Query formulation, e.g. graphical querying
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/40—Processing or translation of natural language
Definitions
- the present invention relates to a search device and search method for searching based on search conditions, and further to a computer-readable recording medium recording programs for realizing these.
- Patent Literature 1 discloses a dialogue scenario generation system that generates scenario data that allows the dialogue to continue until the FAQ data desired by the user is obtained.
- Patent Document 1 is a system that searches for FAQ data desired by the user through interaction between the user and the system, so it is difficult to apply it to searching for image data. That is, it is difficult to apply the system of Patent Literature 1 because there are many similarities and duplications in image data and the like.
- it aims to provide a search device, a search method, and a computer-readable recording medium for efficiently obtaining search results by updating search conditions based on user answers to questions. .
- the search device in one aspect is Using a search condition having one or more attribute information, referring to a data set having search target information in which an image is associated with one or more attribute information, the attribute matching the attribute information of the search condition a search target extraction unit that extracts search target information having information; If the number of the extracted search target information is not within the search result range set in advance, refer to a knowledge base in which attribute information is hierarchically classified, and search in advance for each attribute information included in the extracted search target information.
- a score calculation unit that calculates a score using a determined score function
- a question generation unit that selects attribute information based on the calculated score and generates question information representing a question to be presented to the user using the selected attribute information
- a search condition generation unit that reflects attribute information represented by the user's answer to the question information in the search condition to generate a new search condition
- a search method includes: Using a search condition having one or more attribute information, referring to a data set having search target information in which an image is associated with one or more attribute information, the attribute matching the attribute information of the search condition a search target extraction step of extracting search target information having information; If the number of the extracted search target information is not within the search result range set in advance, refer to a knowledge base in which attribute information is hierarchically classified, and search in advance for each attribute information included in the extracted search target information.
- a score calculation step of calculating a score using a determined score function
- a question generation step of selecting attribute information based on the calculated score and generating question information representing a question to be presented to the user using the selected attribute information
- a search condition generation step of reflecting attribute information represented by the user's answer to the question information in the search condition to generate a new search condition
- a computer-readable recording medium recording a program includes: to the computer, Using a search condition having one or more attribute information, referring to a data set having search target information in which an image is associated with one or more attribute information, the attribute matching the attribute information of the search condition a search target extraction step of extracting search target information having information; If the number of the extracted search target information is not within the search result range set in advance, refer to a knowledge base in which attribute information is hierarchically classified, and search in advance for each attribute information included in the extracted search target information.
- a score calculation step of calculating a score using a determined score function
- a question generation step of selecting attribute information based on the calculated score and generating question information representing a question to be presented to the user using the selected attribute information
- a search condition generation step of reflecting attribute information represented by the user's answer to the question information in the search condition to generate a new search condition
- FIG. 1 is a diagram showing an example of a search device.
- FIG. 2 is a diagram for explaining an example of the knowledge base.
- FIG. 3 is a diagram for explaining the relationship between search target information and attribute information.
- FIG. 4 is a diagram showing an example of a system having a search device.
- FIG. 5 is a diagram for explaining an example of the operation of the search device;
- FIG. 6 is a diagram showing an example of a computer that implements the search device.
- FIG. 7 is a diagram for explaining questions and answers.
- FIG. 8 is a diagram for explaining the relationship between search target information and attribute information.
- FIG. 1 is a diagram showing an example of a search device.
- the search device 10 shown in FIG. 1 updates the search conditions based on the user's answer to the question, and can efficiently obtain search results. Further, as shown in FIG. 1 , the search device 10 has a search target extraction unit 11 , a score calculation unit 12 , a question generation unit 13 and a search condition generation unit 14 . The search device 10 is also connected to a knowledge base 15 via a network or the like.
- the search target extracting unit 11 uses a search condition having one or more attribute information to refer to a data set having search target information in which an image is associated with one or more attribute information, and extracts the attribute information of the search condition. Extracts search target information having attribute information that matches with .
- the search target extraction unit 11 extracts images based on search conditions.
- An image is, for example, an image in which a person or an object is captured.
- the image may be a still image or a moving image. Examples of still images include paintings, graphics, clip arts, and illustrations, and examples of moving images include video images and animations, but the types of images are not limited to these.
- the search condition has one or more pieces of attribute information representing image features.
- the attribute information represents characteristics of an image of a person (hereinafter referred to as a person image), for example, gender, age, color of clothes, glasses, etc. can be considered.
- the search target extraction unit 11 uses gender, age, color of clothes, and eyeglasses to associate with each person image. By referring to one or more pieces of attribute information, a person image having the attribute information of gender, age, color of clothes, and glasses is extracted.
- the attribute information associated with each person image is classified hierarchically.
- the attribute information is hierarchized like the knowledge base shown in FIG.
- FIG. 2 is a diagram for explaining an example of the knowledge base. Details of the knowledge base will be described later.
- the data set has search target information in which a person ID, one or more pieces of attribute information, and a person image are associated.
- FIG. 3 is a diagram for explaining the relationship between search target information and attribute information.
- the data set in FIG. 3 has nine items of search target information.
- attribute information such as age, sex, eyeglasses, and color of clothes are hierarchically classified.
- Age has middle-aged and elderly attribute information and young people in the lower layer.
- Middle-aged and elderly have attribute information of 40's, 50's and 60's in the lower layer.
- Young people have attribute information of twenties and thirties in the lower layer.
- Gender has male and female attribute information in the lower layer. There are no glasses in the lower layer (representing the state of not wearing glasses), corrective glasses (representing wearing glasses to correct vision), and sunglasses (representing wearing sunglasses) Has attribute information.
- the color of the clothes has grey-black, orange, red, yellow, green, blue, and purple attribute information in the lower layer.
- Reds have attribute information of deep reds, dark reds, and light corals in the lower layer.
- Crimson has attribute information of crimson (255, 0, 0), orange-red (255, 69, 0), and crimson (220, 20, 60) in the lower layer.
- Dark red has attribute information of dark red (139, 0, 0), brown (165, 42, 42), and fire brick (178, 34, 34) in the lower layer.
- Light corals have attribute information of light coral (240, 128, 128), Indian red (205, 92, 92) and salmon (250, 128, 114) in the lower layer.
- the three numbers in parentheses represent RGB values.
- the structure of the knowledge base is not limited to the structure shown in Figure 2.
- the knowledge base 15 is provided outside the search device 10 in the example of FIG. 1, it may be provided inside the search device 10. FIG.
- the score calculation unit 12 refers to the knowledge base 15 in which attribute information is hierarchically classified, and calculates the attributes included in the extracted search target information. A score is calculated using a predetermined score function for each piece of information.
- the search result range is preset by the user.
- the search result range is information representing the range of the number of search target information obtained by extraction desired by the user.
- the score function includes a reduced score function and an expanded score function.
- the reduction score function is a function used to reduce the number of search target information when the number of search target information obtained by extraction is larger than the search result range.
- the score calculation unit 12 calculates a score using a reduced score function for attribute information having attribute information (leaf node) in a lower layer (lower layer).
- the reduced score function is, for example, Equation 1 or the like.
- the expansion score function is a function used to expand the number of search target information when the number of search target information obtained by extraction is smaller than the search result range.
- the score calculation unit 12 refers to the knowledge base 15, and calculates the score of the attribute information (node) of the upper layer (higher rank) or the attribute information (node) of the same layer (coordinate) using the expanded score function. .
- the augmented score function is, for example, Equation 2 or the like.
- the element function f1i(o) is a function that returns a larger value as the attribute information of the search condition is input later.
- the reason why a larger value is returned the later the order is, is that the later the order in which the attribute information is input, the higher the uncertainty, so it is better to narrow down.
- the element function f1i(o) is, for example, Equation 3 or the like.
- the element function f1i(o) is used to calculate the value for each piece of attribute information.
- men are 1/2, 30's are 2/3, red is 3/4, and glasses are 4/5 (or ⁇ 1).
- the weighting factor w1 is a value greater than 0 and less than or equal to 1. Note that the weighting factor w1 of the element function f1i(o) is determined by experiment, simulation, or the like.
- the element function f2i(d) is a function that calculates a value for the distance from a node to a leaf node in the knowledge base 15. The greater the distance between the node corresponding to the attribute information and the leaf node, the higher the ambiguity and the higher the possibility of narrowing down.
- the function f2i(d) is, for example, Formula 4.
- the element function f2i(d) is used to calculate the value for each piece of attribute information.
- the weighting factor w2 is a value greater than 0 and less than or equal to 1.
- the weighting factor w2 of the element function f2i(d) is determined through experiments, simulations, or the like.
- the element function f3i(r) is a function for calculating the division ratio r for the extracted attribute information.
- the search target information extracted using the lower layer attribute information is divided as evenly as possible. The more even the split, the better the ability to narrow down.
- the function f3i(r) is, for example, Equation 5.
- the ratio r j of the number of pieces of search target information for grapes to the total number will be described.
- the attribute information is red
- attribute information There are three types of attribute information (crimson, orange-red, crimson) in the lower layer of crimson. There are three attribute information (dark red, brown, firebrick) under the dark red class. There are three attribute information (Light Coral, Indian Red, Salmon) in the lower layer of Light Coral.
- Crimson and Crimson which are lower layers of Crimson, are associated with search target information related to person IDs 1 and 2, the ratio r1 of Crimson to the total number is 2/9. Since dark red, brown, and firebrick, which are lower layers of dark red, are associated with the search target information related to person IDs 3, 4, 5, and 6, the ratio r2 of the total number of dark red is 4 . /9. Since light coral and salmon, which are below light corals, are associated with search target information related to person IDs 7, 8, and 9, the ratio r3 of light corals to the total number is 3/9. Therefore, the element function f3i(r) when the attribute information is red is given by Equation (6).
- the element function f3i(r) is Formula 7.
- the weighting factor w3 is a value greater than 0 and less than or equal to 1.
- the weighting factor w3 of the element function f3i(r) is determined through experiments, simulations, or the like.
- the attribute information of the search condition and the attribute information of the attribute information of the search condition are combined to calculate the similarity, and the maximum value of the calculated similarities is is similarity s.
- the function f4i(s) is, for example, Equation 8.
- the weighting factor w4 is a value greater than 0 and less than or equal to 1. Note that the weighting factor w4 of the element function f4i(s) is determined by experiments, simulations, or the like.
- the question generation unit 13 selects attribute information based on the calculated score, and uses the selected attribute information to generate question information representing questions to be presented to the user. Specifically, first, the question generation unit 13 compares scores calculated for each piece of attribute information, and selects attribute information corresponding to the highest score. Next, the question generation unit 13 generates question information representing a question to be presented to the user using the selected attribute information.
- the search condition generation unit 14 reflects the attribute information included in the user's answer to the question information in the search conditions and generates new search conditions.
- the search conditions are updated based on the user's answer to the question, and the updated search information is used for the search, so the search results can be obtained efficiently.
- FIG. 4 is a diagram showing an example of a system having a search device.
- the system 40 in the embodiment has a search device 10, a knowledge base 15, and an input/output device 41.
- the search device 10 is, for example, a CPU (Central Processing Unit), a programmable device such as an FPGA (Field-Programmable Gate Array), or a GPU (Graphics Processing Unit), or one or more of them.
- Information processing devices such as circuits, server computers, personal computers, and mobile terminals.
- the input/output device 41 has a user interface, and has an input section for the user to input information and an output section for outputting images and sounds to the user.
- the input unit is, for example, an input device having a keyboard, mouse, touch panel, and the like.
- the output unit is, for example, an image display device using liquid crystal, organic EL (Electro Luminescence), or CRT (Cathode Ray Tube).
- the image display device may include an audio output device such as a speaker.
- the output unit may be a printing device such as a printer.
- the search target extraction unit 11, the score calculation unit 12, the question generation unit 13, the search condition generation unit 14, and the knowledge base 15 of the search device 10 have already been described, so descriptions thereof will be omitted.
- FIG. 5 is a diagram for explaining an example of the operation of the search device; In the following description, reference will be made to the drawings as appropriate. Also, in the embodiment, the search method is implemented by operating the search device. Therefore, the description of the search method in the embodiment is replaced with the description of the operation of the search device below.
- step A2 the search target extraction unit 11 acquires initial search conditions at the start of search (step A2). Specifically, in step A2 at the start of search, the document information created by the user is analyzed using a known document analysis tool or the like, and attribute information of search conditions is acquired. For example, a document such as "Gender is male, age is about 30, and the color of the clothes is red” is acquired, this document is analyzed, and the attribute information of the search conditions is sex: male, age: 30s, clothes color: get red and so on.
- step A2 after the search condition is updated, the search target extraction unit 11 acquires the updated search information.
- the search target extracting unit 11 uses the attribute information of the search condition to refer to the attribute information associated with the search target information, and searches for search target information that matches any one or more of the attribute information of the search condition. Extract and set the search target information (data set) (step A3).
- the search target extracting unit 11 uses a search condition having one or more attribute information to refer to a data set having search target information in which an image is associated with one or more attribute information, and extracts the search condition (step A4).
- the score calculation unit 12 determines whether or not the number of pieces of search target information is within a preset search result range (step A5). If the number of search target information is not within the search result range, refer to the knowledge base where attribute information is classified hierarchically, and use a predetermined score function for each attribute information included in the extracted search target information. Calculate the score.
- step A5 calculate reduced score
- the score calculation unit 12 determines that attribute information (leaf node) is in the lower layer (lower).
- a score is calculated for existing attribute information using a reduced score function (step A6).
- the reduced score function is, for example, Equation 1 or the like.
- the question generation unit 13 selects attribute information based on the calculated score, and generates question information representing a question to be presented to the user using the selected attribute information (step A7). Specifically, in step A7, the question generation unit 13 compares the scores calculated for each piece of attribute information in step A6, and selects the attribute information corresponding to the highest score. Next, the question generation unit 13 generates question information representing a question to be presented to the user using the selected attribute information.
- the score calculation unit 12 calculates the attribute information (node) of the upper layer (higher rank) or the same layer (same rank ) for the attribute information (node), the score is calculated using the extended score function (step A8).
- the augmented score function is, for example, Equation 2 or the like.
- the question generation unit 13 selects attribute information based on the calculated score, and generates question information representing a question to be presented to the user using the selected attribute information (step A9). Specifically, in step A9, the question generation unit 13 compares the scores calculated for each piece of attribute information in step A8, and selects the attribute information corresponding to the highest score. Next, the question generation unit 13 generates question information representing a question to be presented to the user using the selected attribute information.
- the question generation unit 13 outputs the question information to the input/output device 41 (step A10).
- the search condition generator 14 acquires answer information representing the user's answer to the question information (step A11).
- the search condition generator 14 determines whether or not to update the search conditions (step A12). If the search conditions are to be updated (step A12: Yes), the search condition generator 14 reflects the attribute information in the search conditions to generate new search conditions (step A13). If the search condition is not to be updated (step A12: No), the process proceeds to step A5 to generate question information again.
- the search condition generator 14 updates the data set (step A14). Specifically, in step A14, the search condition generator 14 extracts search target information having the attribute information used in updating the search conditions, and uses the extracted search target information as a new data set.
- step A2 the search is continued using the updated search conditions and the updated data set. Then, if the number of pieces of information to be searched is within the search result range, the search processing is terminated (step A5: end of search).
- the search condition is updated based on the user's answer to the question, and the updated search information is used for the search, so that the search result can be obtained efficiently.
- the user can also more accurately grasp the attribute information of the search target.
- the program in the embodiment may be any program that causes a computer to execute steps A1 to A14 shown in FIG.
- the processor of the computer functions as a search target extraction unit 11, a score calculation unit 12, a question generation unit 13, and a search condition generation unit 14, and performs processing.
- each computer may function as one of the search target extractor 11, the score calculator 12, the question generator 13, and the search condition generator 14, respectively.
- FIG. 6 is a diagram showing an example of a computer that implements the search device in the embodiment.
- the computer 110 includes a CPU (Central Processing Unit) 111, a main memory 112, a storage device 113, an input interface 114, a display controller 115, a data reader/writer 116, and a communication interface 117. and These units are connected to each other via a bus 121 so as to be able to communicate with each other.
- the computer 110 may include a GPU or FPGA in addition to or instead of the CPU 111 .
- the CPU 111 expands the programs (codes) in this embodiment stored in the storage device 113 into the main memory 112 and executes them in a predetermined order to perform various calculations.
- the main memory 112 is typically a volatile storage device such as DRAM (Dynamic Random Access Memory).
- the program in this embodiment is provided in a state stored in a computer-readable recording medium 120 .
- the program in this embodiment may be distributed on the Internet connected via the communication interface 117 .
- the recording medium 120 is a non-volatile recording medium.
- the storage device 113 includes hard disk drives and semiconductor storage devices such as flash memory.
- Input interface 114 mediates data transmission between CPU 111 and input devices 118 such as a keyboard and mouse.
- the display controller 115 is connected to the display device 119 and controls display on the display device 119 .
- the data reader/writer 116 mediates data transmission between the CPU 111 and the recording medium 120, reads programs from the recording medium 120, and writes processing results in the computer 110 to the recording medium 120.
- Communication interface 117 mediates data transmission between CPU 111 and other computers.
- the recording medium 120 include general-purpose semiconductor storage devices such as CF (Compact Flash (registered trademark)) and SD (Secure Digital), magnetic recording media such as flexible disks, and CD- Optical recording media such as ROM (Compact Disk Read Only Memory) can be mentioned.
- CF Compact Flash
- SD Secure Digital
- magnetic recording media such as flexible disks
- CD- Optical recording media such as ROM (Compact Disk Read Only Memory) can be mentioned.
- search device 10 in the embodiment can also be realized by using hardware corresponding to each part instead of a computer in which a program is installed. Furthermore, the search device 10 may be partly implemented by a program and the rest by hardware.
- the search target extraction unit 11 acquires search conditions (step A2). For example, sex: male, age: 30's, and color of clothes: red are acquired as attribute information of search conditions.
- the search target extracting unit 11 refers to the attribute information of each of the search target information in the data set using male, 30s, and red as the attribute information of the search condition, and extracts the attribute information matching the attribute information of the search condition. (step A4). In the embodiment, it is assumed that the search target information shown in FIG. 3 is extracted.
- search target information that matches the search conditions male, 30s, red
- search target information corresponding to person image IDs 1, 2, and 7 is selected. Also, the number of pieces of extracted search target information is three.
- the score calculation unit 12 determines whether or not the number of pieces of extracted search target information is within the search result range (step A5). Since the number of extracted search target information is 3, it is larger than 2 set in the search result range (3>2). Therefore, the reduced score is calculated (step A5: calculate reduced score).
- the score calculation unit 12 refers to the knowledge base 15 and calculates the reduced score of the attribute information in the lower nodes (step A6).
- the score shown in Equation 10 is calculated using, for example, the reduced score function S1(i) in Equation 9.
- the question generation unit 13 selects attribute information with the highest score from among the calculated scores. Since red has the highest score in Equation 10, red attribute information is selected.
- the question generation unit 13 acquires the attribute information of bright reds, dark reds, and light corals that are below the attribute information of red. Then, the question generation unit 13 presents question information for displaying on the output unit, for example, "Which red color is closest to bright red, dark red, or light coral?" is generated (step A7).
- the question generation unit 13 outputs the question information to the output device and presents the question to the user (step A10).
- the question is displayed as shown in question display 71 in FIG. However, it is not limited to the question display 71 .
- FIG. 7 is a diagram for explaining questions and answers.
- the search condition generator 14 acquires answer information representing the user's answer to the question information (step A11).
- the user answers dark red.
- the answers are displayed as shown in answer display 72 in FIG. However, it is not limited to the answer display 72 .
- step A12 determines whether to update the search conditions. Then, if the search condition is to be updated (step A12: Yes), the search condition generator 14 updates the red of the search condition to dark red (step A13). If the search condition is not to be updated (step A12: No), the process proceeds to step A5 to generate question information again and present the question.
- the search condition generator 14 updates the data set (step A14).
- search target information that does not have dark red is excluded.
- the data set is only search target information corresponding to person IDs 3 to 6.
- the search target extraction unit 11 acquires the updated search conditions.
- Gender Male
- Age 30's
- Clothes Color Dark Red are acquired as attribute information of the updated search condition.
- the search target extraction unit 11 sets the updated data set (step A3).
- the search target extraction unit 11 extracts search target information having attribute information that matches the updated search conditions (male, 30s, dark red) from the updated data set (step A4). However, since no search target information is extracted, the number of extracted search target information is zero.
- the score calculation unit 12 determines whether or not the number of pieces of extracted search target information is within the search result range (step A5). Since the number of extracted search target information is 0, it is smaller than 1 set in the search result range (0 ⁇ 1), so an expanded score is calculated (step A5: calculate expanded score).
- the score calculation unit 12 refers to the knowledge base 15 and calculates the expanded score of the attribute information in the higher or equivalent node (step A8).
- the score is calculated using, for example, the expanded score function S2(i) of Equation 11.
- the element function f4i(s) is calculated.
- the similarity is calculated by combining the attribute information of the search condition and the attribute information at the higher level or the same node of the attribute information of the search condition.
- the maximum similarity value of women in the same rank as men is 0.2
- the maximum similarity of women in their 40s (40s expanded from 30s) in the same rank as their 30s. is 0.8.
- the question generation unit 13 selects attribute information with the highest score from among the calculated scores. Since the score of people in their 40s is the highest in Equation 12, the attribute information of people in their 40s is selected.
- the question generation unit 13 acquires attribute information for people in their 40s. Then, the question generation unit 13 generates question information to be presented to the user, for example, "Are you in your 40s?" (step A9).
- the question generation unit 13 outputs the question information to the output device and presents the question to the user (step A10).
- the question is displayed as shown in question display 73 in FIG. However, it is not limited to the question display 73 .
- the search condition generator 14 acquires answer information representing the user's answer to the question information (step A11).
- answer information representing the user's answer to the question information
- the answers are displayed as shown in answer display 74 in FIG. However, it is not limited to the answer display 74 .
- step A12 determines whether to update the search conditions. Then, if the search condition is to be updated (step A12: Yes), the search condition generator 14 updates the search condition by adding people in their 40s (step A13). If the search condition is not updated (step A12: No), the process proceeds to step A5 and the question is presented again.
- the search condition generator 14 updates the data set (step A14).
- search target information for people in their 40s is added. However, since there are already people in their 40s, they will not be updated.
- the search target extraction unit 11 acquires the updated search conditions.
- gender male, age: 30's or 40's, and clothing color: dark red are acquired as attribute information of the updated search condition.
- the search target extraction unit 11 sets a data set (step A3).
- the search target extraction unit 11 extracts search target information that matches the search conditions (male, 30's or 40's, dark red) from the data set (step A4).
- person image IDs 4, 5 and 6 are selected.
- the number of pieces of extracted search target information is three.
- the score calculation unit 12 determines whether or not the number of pieces of extracted search target information is within the search result range (step A5). Since the number of pieces of search target information extracted is 3, it is larger than 2 set in the search result range (3>2). Therefore, the reduced score is calculated (step A5: calculate reduced score).
- the score calculation unit 12 refers to the knowledge base 15 and calculates reduced scores for the attributes in the lower nodes (step A6).
- the element functions f3i(r) for each of dark red and eyeglasses are calculated.
- the number of data is 4, the number of lower nodes is 3, and the standard division is 2, 1, 1. Therefore, the element functions f3i(r) for dark red and glasses are given by Equation 13.
- Equation 14 a reduced score function S1(i) is calculated for each piece of attribute information as shown in Equation 14.
- the question generation unit 13 selects attribute information with the highest score from among the calculated scores.
- glasses have the highest score, so the attribute information of glasses is selected.
- the question generation unit 13 acquires the attribute information of the glasses. Then, the question generation unit 13 generates question information to be presented to the user, for example, "What kind of glasses do you wear? No, corrective glasses, sunglasses" (step A9).
- the question generation unit 13 outputs the question information to the output device and presents the question to the user (step A10).
- the question is displayed as shown in question display 75 in FIG. However, it is not limited to the question display 75 .
- the search condition generator 14 acquires answer information representing the user's answer to the question information (step A11).
- the user answers "sunglasses".
- the answers are displayed as shown in answer display 76 in FIG. However, it is not limited to the answer display 76 .
- step A12 determines whether to update the search conditions. Then, if the search condition is to be updated (step A12: Yes), the search condition generator 14 adds sunglasses to the search condition (step A13). If the search condition is not to be updated (step A12: No), the process proceeds to step A5 to generate question information again and present the question.
- the search condition generator 14 updates the data set (step A14).
- search target information that does not have sunglasses is excluded.
- the data set is only the search target information corresponding to person ID 6 shown in FIG.
- FIG. 8 is a diagram for explaining the relationship between search target information and attribute information.
- the search target extraction unit 11 acquires the updated search conditions.
- gender male, age: 30's or 40's, color of clothes: dark red, glasses: sunglasses are acquired as attribute information of updated search conditions.
- the search target extraction unit 11 sets a data set (step A3).
- the search target extraction unit 11 extracts search target information that matches the search conditions (male, 30's or 40's, dark red, sunglasses) from the data set (step A4).
- person image ID6 is selected.
- the number of pieces of extracted search target information is one.
- a search device having
- the search device refers to the knowledge base and uses a reduction score function to reduce the search results for each attribute information of a lower node when the number of the extracted search target information is larger than the search result range. calculate a score, search device.
- the search device according to any one of Appendices 1 to 3,
- the search target information is information in which identification information for identifying the search target information, the attribute information, and a person image related to the search target information are associated with each other. search device.
- a score calculation step of calculating a score using a determined score function
- a question generation step of selecting attribute information based on the calculated score and generating question information representing a question to be presented to the user using the selected attribute information
- a search condition generation step of reflecting attribute information represented by the user's answer to the question information in the search condition to generate a new search condition
- Appendix 7 The search method according to appendix 5 or 6, In the score calculation step, when the number of the extracted search target information is larger than the search result range, the knowledge base is referred to, and a reduction score function is used to reduce the search results for each attribute information of lower nodes. calculate a score, retrieval method.
- the search target information is information in which identification information for identifying the search target information, the attribute information, and a person image related to the search target information are associated with each other. retrieval method.
- a computer-readable recording medium recording a program containing instructions for executing a
- Appendix 10 The computer-readable recording medium according to Appendix 9, In the score calculation step, when the number of the extracted search target information is smaller than the search result range, an expansion score function for referring to the knowledge base and expanding the search results for each attribute information of a higher node or a peer node. Calculate the score using Computer-readable recording medium.
- Appendix 11 The computer-readable recording medium according to Appendix 9 or 10, In the score calculation step, when the number of the extracted search target information is larger than the search result range, the knowledge base is referred to, and a reduction score function is used to reduce the search results for each attribute information of lower nodes. calculate a score, Computer-readable recording medium.
- the computer-readable recording medium according to any one of appendices 9 to 11,
- the search target information is information in which identification information for identifying the search target information, the attribute information, and a person image related to the search target information are associated with each other.
- Computer-readable recording medium is information in which identification information for identifying the search target information, the attribute information, and a person image related to the search target information are associated with each other.
- the present invention it is possible to efficiently obtain search results by updating the search conditions based on the user's answer to the question.
- INDUSTRIAL APPLICABILITY The present invention is useful in fields where it is necessary to perform searches efficiently.
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Mathematical Physics (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- General Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
Description
一つ以上の属性情報を有する検索条件を用いて、画像と一つ以上の属性情報とが関連付けられた検索対象情報を有するデータセットを参照し、前記検索条件の前記属性情報と一致する前記属性情報を有する検索対象情報を抽出する、検索対象抽出部と、
抽出した前記検索対象情報の数があらかじめ設定された検索結果範囲にない場合、属性情報が階層的に分類されたナレッジベースを参照し、抽出した前記検索対象情報に含まれる属性情報ごとに、あらかじめ決められたスコア関数を用いてスコアを算出する、スコア算出部と、
算出した前記スコアに基づいて属性情報を選択し、選択した前記属性情報を用いて利用者に提示する質問を表す質問情報を生成する、質問生成部と、
前記質問情報に対する前記利用者の回答が表す属性情報を前記検索条件に反映させ、新たな検索条件を生成する、検索条件生成部と、
を有することを特徴とする。
一つ以上の属性情報を有する検索条件を用いて、画像と一つ以上の属性情報とが関連付けられた検索対象情報を有するデータセットを参照し、前記検索条件の前記属性情報と一致する前記属性情報を有する検索対象情報を抽出する、検索対象抽出ステップと、
抽出した前記検索対象情報の数があらかじめ設定された検索結果範囲にない場合、属性情報が階層的に分類されたナレッジベースを参照し、抽出した前記検索対象情報に含まれる属性情報ごとに、あらかじめ決められたスコア関数を用いてスコアを算出する、スコア算出ステップと、
算出した前記スコアに基づいて属性情報を選択し、選択した前記属性情報を用いて利用者に提示する質問を表す質問情報を生成する、質問生成ステップと、
前記質問情報に対する前記利用者の回答が表す属性情報を前記検索条件に反映させ、新たな検索条件を生成する、検索条件生成ステップと、
を有することを特徴とする。
コンピュータに、
一つ以上の属性情報を有する検索条件を用いて、画像と一つ以上の属性情報とが関連付けられた検索対象情報を有するデータセットを参照し、前記検索条件の前記属性情報と一致する前記属性情報を有する検索対象情報を抽出する、検索対象抽出ステップと、
抽出した前記検索対象情報の数があらかじめ設定された検索結果範囲にない場合、属性情報が階層的に分類されたナレッジベースを参照し、抽出した前記検索対象情報に含まれる属性情報ごとに、あらかじめ決められたスコア関数を用いてスコアを算出する、スコア算出ステップと、
算出した前記スコアに基づいて属性情報を選択し、選択した前記属性情報を用いて利用者に提示する質問を表す質問情報を生成する、質問生成ステップと、
前記質問情報に対する前記利用者の回答が表す属性情報を前記検索条件に反映させ、新たな検索条件を生成する、検索条件生成ステップと、
を実行させるプログラムを記録していることを特徴とする。
図1を用いて、実施形態における検索装置10の構成について説明する。図1は、検索装置の一例を示す図である。
図1に示す検索装置10は、質問に対する利用者の回答に基づいて検索条件を更新して、検索結果を効率よく取得できる。また、図1に示すように、検索装置10は、検索対象抽出部11と、スコア算出部12と、質問生成部13と、検索条件生成部14とを有する。また、検索装置10は、ネットワークなどを介してナレッジベース15と接続されている。
S1(i)=(w1×f1i(o))+(w2×f2i(d))+(w3×f3i(r))
S1(i):縮小スコア
i :属性情報を識別する識別子
w1 :要素関数f1i(o)の重み係数
f1i(o):属性情報が入力される順番oに対する値を算出する関数
w2 :要素関数f2i(d)の重み係数
f2i(d):葉ノードとの距離dに対する値を算出する関数
w3 :要素関数f3i(r)の重み係数
f3i(r):抽出された属性情報に対する分割率rを算出する関数
S2(i)=(w1×f1i(o))+(w4×f4i(s))
S2(i):拡大スコア
w4 :要素関数f4i(s)の重み係数
f4i(s):属性情報の類似度sに対する値を算出する関数
f1i(o)=1-1/(o+1)
o :属性情報の入力の順番
f2i(d)=1-1/(d+1)
d :ノードと葉ノードとの距離
f4i(s)=s=maxSim(an,am)
n、m:属性情報を識別する識別子
an :検索条件の属性情報
am :検索条件の属性情報の上位又は同位ノードにある属性情報
Sim(an,ai):属性情報の類似度を算出する関数(例えば、cosine関数)
図4を用いて、実施形態における検索装置10の構成をより具体的に説明する。図4は、検索装置を有するシステムの一例を示す図である。
実施形態における検索装置の動作について図5を用いて説明する。図5は、検索装置の動作の一例を説明するための図である。以下の説明においては、適宜図を参照する。また、実施形態では、検索装置を動作させることによって、検索方法が実施される。よって、実施形態における検索方法の説明は、以下の検索装置の動作説明に代える。
以上のように実施形態によれば、質問に対する利用者の回答に基づいて検索条件を更新し、更新した検索情報を用いて検索をするので、検索結果を効率よく取得することができる。質問に対する回答と検索条件の更新により、利用者は検索対象の属性情報をより正確に把握することもできる。
実施形態におけるプログラムは、コンピュータに、図5に示すステップA1からA14を実行させるプログラムであればよい。このプログラムをコンピュータにインストールし、実行することによって、実施形態における検索装置と検索方法とを実現することができる。この場合、コンピュータのプロセッサは、検索対象抽出部11、スコア算出部12、質問生成部13、検索条件生成部14として機能し、処理を行なう。
ここで、実施形態におけるプログラムを実行することによって、検索装置を実現するコンピュータについて図6を用いて説明する。図6は、実施形態における検索装置を実現するコンピュータの一例を示す図である。
S1(i)=(0.3×f1i(o))+(0.3×f2i(d))+(0.4×f3i(r))
S1(赤) =0.3×3/4+0.3×2/3+0.4×7/9≒0.726
S1(眼鏡)=0.3×1 +0.3×1/2+0.4×3/9≒0.583
・・・
S2(i)=(0.5×f1i(o))+(0.5×f4i(s))
S2(女性) =0.5×1/2+0.5×0.2=0.35
S2(40代)=0.5×2/3+0.5×0.8≒0.734
S1(暗赤)=0.3×3/4+0.3×1/2+0.4×1=0.775
S1(眼鏡)=0.3×1 +0.3×1/2+0.4×1=0.85
・・・
以上の実施形態に関し、更に以下の付記を開示する。上述した実施形態の一部又は全部は、以下に記載する(付記1)から(付記12)により表現することができるが、以下の記載に限定されるものではない。
一つ以上の属性情報を有する検索条件を用いて、画像と一つ以上の属性情報とが関連付けられた検索対象情報を有するデータセットを参照し、前記検索条件の前記属性情報と一致する前記属性情報を有する検索対象情報を抽出する、検索対象抽出部と、
抽出した前記検索対象情報の数があらかじめ設定された検索結果範囲にない場合、属性情報が階層的に分類されたナレッジベースを参照し、抽出した前記検索対象情報に含まれる属性情報ごとに、あらかじめ決められたスコア関数を用いてスコアを算出する、スコア算出部と、
算出した前記スコアに基づいて属性情報を選択し、選択した前記属性情報を用いて利用者に提示する質問を表す質問情報を生成する、質問生成部と、
前記質問情報に対する前記利用者の回答が表す属性情報を前記検索条件に反映させ、新たな検索条件を生成する、検索条件生成部と、
を有する検索装置。
付記1に記載の検索装置であって、
前記スコア算出部は、抽出した前記検索対象情報の数が前記検索結果範囲より小さい場合、前記ナレッジベースを参照し、上位ノード又は同位ノードの属性情報ごとに検索結果を拡大するための拡大スコア関数を用いてスコアを算出する、
検索装置。
付記1又は2に記載の検索装置であって、
前記スコア算出部は、抽出した前記検索対象情報の数が前記検索結果範囲より大きい場合、前記ナレッジベースを参照し、下位ノードの属性情報ごとに検索結果を縮小するための縮小スコア関数を用いてスコアを算出する、
検索装置。
付記1から3のいずれか一つに記載の検索装置であって、
前記検索対象情報は、前記検索対象情報を識別する識別情報と、前記属性情報と、前記検索対象情報に関連する人物画像とが関連付けられた情報である、
検索装置。
一つ以上の属性情報を有する検索条件を用いて、画像と一つ以上の属性情報とが関連付けられた検索対象情報を有するデータセットを参照し、前記検索条件の前記属性情報と一致する前記属性情報を有する検索対象情報を抽出する、検索対象抽出ステップと、
抽出した前記検索対象情報の数があらかじめ設定された検索結果範囲にない場合、属性情報が階層的に分類されたナレッジベースを参照し、抽出した前記検索対象情報に含まれる属性情報ごとに、あらかじめ決められたスコア関数を用いてスコアを算出する、スコア算出ステップと、
算出した前記スコアに基づいて属性情報を選択し、選択した前記属性情報を用いて利用者に提示する質問を表す質問情報を生成する、質問生成ステップと、
前記質問情報に対する前記利用者の回答が表す属性情報を前記検索条件に反映させ、新たな検索条件を生成する、検索条件生成ステップと、
を有する検索方法。
付記5に記載の検索方法であって、
前記スコア算出ステップにおいて、抽出した前記検索対象情報の数が前記検索結果範囲より小さい場合、前記ナレッジベースを参照し、上位ノード又は同位ノードの属性情報ごとに検索結果を拡大するための拡大スコア関数を用いてスコアを算出する、
検索方法。
付記5又は6に記載の検索方法であって、
前記スコア算出ステップにおいて、抽出した前記検索対象情報の数が前記検索結果範囲より大きい場合、前記ナレッジベースを参照し、下位ノードの属性情報ごとに検索結果を縮小するための縮小スコア関数を用いてスコアを算出する、
検索方法。
付記5から7のいずれか一つに記載の検索方法であって、
前記検索対象情報は、前記検索対象情報を識別する識別情報と、前記属性情報と、前記検索対象情報に関連する人物画像とが関連付けられた情報である、
検索方法。
コンピュータに、
一つ以上の属性情報を有する検索条件を用いて、画像と一つ以上の属性情報とが関連付けられた検索対象情報を有するデータセットを参照し、前記検索条件の前記属性情報と一致する前記属性情報を有する検索対象情報を抽出する、検索対象抽出ステップと、
抽出した前記検索対象情報の数があらかじめ設定された検索結果範囲にない場合、属性情報が階層的に分類されたナレッジベースを参照し、抽出した前記検索対象情報に含まれる属性情報ごとに、あらかじめ決められたスコア関数を用いてスコアを算出する、スコア算出ステップと、
算出した前記スコアに基づいて属性情報を選択し、選択した前記属性情報を用いて利用者に提示する質問を表す質問情報を生成する、質問生成ステップと、
前記質問情報に対する前記利用者の回答が表す属性情報を前記検索条件に反映させ、新たな検索条件を生成する、検索条件生成ステップと、
を実行させる命令を含む、プログラムを記録しているコンピュータ読み取り可能な記録媒体。
付記9に記載のコンピュータ読み取り可能な記録媒体であって、
前記スコア算出ステップにおいて、抽出した前記検索対象情報の数が前記検索結果範囲より小さい場合、前記ナレッジベースを参照し、上位ノード又は同位ノードの属性情報ごとに検索結果を拡大するための拡大スコア関数を用いてスコアを算出する、
コンピュータ読み取り可能な記録媒体。
付記9又は10に記載のコンピュータ読み取り可能な記録媒体であって、
前記スコア算出ステップにおいて、抽出した前記検索対象情報の数が前記検索結果範囲より大きい場合、前記ナレッジベースを参照し、下位ノードの属性情報ごとに検索結果を縮小するための縮小スコア関数を用いてスコアを算出する、
コンピュータ読み取り可能な記録媒体。
付記9から11のいずれか一つに記載のコンピュータ読み取り可能な記録媒体であって、
前記検索対象情報は、前記検索対象情報を識別する識別情報と、前記属性情報と、前記検索対象情報に関連する人物画像とが関連付けられた情報である、
コンピュータ読み取り可能な記録媒体。
11 検索対象抽出部
12 スコア算出部
13 質問生成部
14 検索条件生成部
15 ナレッジベース
40 システム
41 入出力装置
110 コンピュータ
111 CPU
112 メインメモリ
113 記憶装置
114 入力インターフェイス
115 表示コントローラ
116 データリーダ/ライタ
117 通信インターフェイス
118 入力機器
119 ディスプレイ装置
120 記録媒体
121 バス
Claims (12)
- 一つ以上の属性情報を有する検索条件を用いて、画像と一つ以上の属性情報とが関連付けられた検索対象情報を有するデータセットを参照し、前記検索条件の前記属性情報と一致する前記属性情報を有する検索対象情報を抽出する、検索対象抽出手段と、
抽出した前記検索対象情報の数があらかじめ設定された検索結果範囲にない場合、属性情報が階層的に分類されたナレッジベースを参照し、抽出した前記検索対象情報に含まれる属性情報ごとに、あらかじめ決められたスコア関数を用いてスコアを算出する、スコア算出手段と、
算出した前記スコアに基づいて属性情報を選択し、選択した前記属性情報を用いて利用者に提示する質問を表す質問情報を生成する、質問生成手段と、
前記質問情報に対する前記利用者の回答が表す属性情報を前記検索条件に反映させ、新たな検索条件を生成する、検索条件生成手段と、
を有する検索装置。 - 請求項1に記載の検索装置であって、
前記スコア算出手段は、抽出した前記検索対象情報の数が前記検索結果範囲より小さい場合、前記ナレッジベースを参照し、上位ノード又は同位ノードの属性情報ごとに検索結果を拡大するための拡大スコア関数を用いてスコアを算出する、
検索装置。 - 請求項1又は2に記載の検索装置であって、
前記スコア算出手段は、抽出した前記検索対象情報の数が前記検索結果範囲より大きい場合、前記ナレッジベースを参照し、下位ノードの属性情報ごとに検索結果を縮小するための縮小スコア関数を用いてスコアを算出する、
検索装置。 - 請求項1から3のいずれか一つに記載の検索装置であって、
前記検索対象情報は、前記検索対象情報を識別する識別情報と、前記属性情報と、前記検索対象情報に関連する人物画像とが関連付けられた情報である、
検索装置。 - 一つ以上の属性情報を有する検索条件を用いて、画像と一つ以上の属性情報とが関連付けられた検索対象情報を有するデータセットを参照し、前記検索条件の前記属性情報と一致する前記属性情報を有する検索対象情報を抽出し、
抽出した前記検索対象情報の数があらかじめ設定された検索結果範囲にない場合、属性情報が階層的に分類されたナレッジベースを参照し、抽出した前記検索対象情報に含まれる属性情報ごとに、あらかじめ決められたスコア関数を用いてスコアを算出し、
算出した前記スコアに基づいて属性情報を選択し、選択した前記属性情報を用いて利用者に提示する質問を表す質問情報を生成し、
前記質問情報に対する前記利用者の回答が表す属性情報を前記検索条件に反映させ、新たな検索条件を生成する、
検索方法。 - 請求項5に記載の検索方法であって、
前記スコアの算出において、抽出した前記検索対象情報の数が前記検索結果範囲より小さい場合、前記ナレッジベースを参照し、上位ノード又は同位ノードの属性情報ごとに検索結果を拡大するための拡大スコア関数を用いてスコアを算出する、
検索方法。 - 請求項5又は6に記載の検索方法であって、
前記スコアの算出において、抽出した前記検索対象情報の数が前記検索結果範囲より大きい場合、前記ナレッジベースを参照し、下位ノードの属性情報ごとに検索結果を縮小するための縮小スコア関数を用いてスコアを算出する、
検索方法。 - 請求項5から7のいずれか一つに記載の検索方法であって、
前記検索対象情報は、前記検索対象情報を識別する識別情報と、前記属性情報と、前記検索対象情報に関連する人物画像とが関連付けられた情報である、
検索方法。 - コンピュータに、
一つ以上の属性情報を有する検索条件を用いて、画像と一つ以上の属性情報とが関連付けられた検索対象情報を有するデータセットを参照し、前記検索条件の前記属性情報と一致する前記属性情報を有する検索対象情報を抽出させ、
抽出した前記検索対象情報の数があらかじめ設定された検索結果範囲にない場合、属性情報が階層的に分類されたナレッジベースを参照し、抽出した前記検索対象情報に含まれる属性情報ごとに、あらかじめ決められたスコア関数を用いてスコアを算出させ、
算出した前記スコアに基づいて属性情報を選択し、選択した前記属性情報を用いて利用者に提示する質問を表す質問情報を生成させ、
前記質問情報に対する前記利用者の回答が表す属性情報を前記検索条件に反映させ、新たな検索条件を生成させる、
命令を含む、プログラムを記録しているコンピュータ読み取り可能な記録媒体。 - 請求項9に記載のコンピュータ読み取り可能な記録媒体であって、
前記スコアの算出において、抽出した前記検索対象情報の数が前記検索結果範囲より小さい場合、前記ナレッジベースを参照し、上位ノード又は同位ノードの属性情報ごとに検索結果を拡大するための拡大スコア関数を用いてスコアを算出させる、
コンピュータ読み取り可能な記録媒体。 - 請求項9又は10に記載のコンピュータ読み取り可能な記録媒体であって、
前記スコアの算出において、抽出した前記検索対象情報の数が前記検索結果範囲より大きい場合、前記ナレッジベースを参照し、下位ノードの属性情報ごとに検索結果を縮小するための縮小スコア関数を用いてスコアを算出させる、
コンピュータ読み取り可能な記録媒体。 - 請求項9から11のいずれか一つに記載のコンピュータ読み取り可能な記録媒体であって、
前記検索対象情報は、前記検索対象情報を識別する識別情報と、前記属性情報と、前記検索対象情報に関連する人物画像とが関連付けられた情報である、
ことを特徴とするコンピュータ読み取り可能な記録媒体。
Priority Applications (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/JP2021/013097 WO2022201541A1 (ja) | 2021-03-26 | 2021-03-26 | 検索装置、検索方法、及びコンピュータ読み取り可能な記録媒体 |
US18/278,762 US12118024B2 (en) | 2021-03-26 | 2021-03-26 | Search apparatus, search method, and computer readable recording medium |
JP2023508410A JPWO2022201541A5 (ja) | 2021-03-26 | 検索装置、検索方法、及びプログラム |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/JP2021/013097 WO2022201541A1 (ja) | 2021-03-26 | 2021-03-26 | 検索装置、検索方法、及びコンピュータ読み取り可能な記録媒体 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2022201541A1 true WO2022201541A1 (ja) | 2022-09-29 |
Family
ID=83396733
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2021/013097 WO2022201541A1 (ja) | 2021-03-26 | 2021-03-26 | 検索装置、検索方法、及びコンピュータ読み取り可能な記録媒体 |
Country Status (2)
Country | Link |
---|---|
US (1) | US12118024B2 (ja) |
WO (1) | WO2022201541A1 (ja) |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH10260990A (ja) * | 1997-03-18 | 1998-09-29 | Nippon Telegr & Teleph Corp <Ntt> | 例示検索の高速化方法 |
JP2007299159A (ja) * | 2006-04-28 | 2007-11-15 | Toshiba Corp | コンテンツ検索装置 |
JP2011107887A (ja) * | 2009-11-16 | 2011-06-02 | Toshiba Corp | データベース装置およびデータベース検索方法 |
Family Cites Families (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6185560B1 (en) * | 1998-04-15 | 2001-02-06 | Sungard Eprocess Intelligance Inc. | System for automatically organizing data in accordance with pattern hierarchies therein |
EP1394727B1 (en) * | 2002-08-30 | 2011-10-12 | MVTec Software GmbH | Hierarchical component based object recognition |
US20110188759A1 (en) * | 2003-06-26 | 2011-08-04 | Irina Filimonova | Method and System of Pre-Analysis and Automated Classification of Documents |
US7545809B2 (en) * | 2003-05-28 | 2009-06-09 | International Business Machines Corporation | Packet classification |
US8484245B2 (en) * | 2011-02-08 | 2013-07-09 | Xerox Corporation | Large scale unsupervised hierarchical document categorization using ontological guidance |
JP2012248161A (ja) | 2011-05-31 | 2012-12-13 | Oki Electric Ind Co Ltd | 対話型検索システム及びプログラム、並びに、対話シナリオ生成システム及びプログラム |
US9928448B1 (en) * | 2016-09-23 | 2018-03-27 | International Business Machines Corporation | Image classification utilizing semantic relationships in a classification hierarchy |
US11417082B2 (en) * | 2017-06-16 | 2022-08-16 | Markable, Inc. | Image processing system |
GB2586265B (en) * | 2019-08-15 | 2023-02-15 | Vision Semantics Ltd | Text based image search |
-
2021
- 2021-03-26 WO PCT/JP2021/013097 patent/WO2022201541A1/ja active Application Filing
- 2021-03-26 US US18/278,762 patent/US12118024B2/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH10260990A (ja) * | 1997-03-18 | 1998-09-29 | Nippon Telegr & Teleph Corp <Ntt> | 例示検索の高速化方法 |
JP2007299159A (ja) * | 2006-04-28 | 2007-11-15 | Toshiba Corp | コンテンツ検索装置 |
JP2011107887A (ja) * | 2009-11-16 | 2011-06-02 | Toshiba Corp | データベース装置およびデータベース検索方法 |
Also Published As
Publication number | Publication date |
---|---|
JPWO2022201541A1 (ja) | 2022-09-29 |
US12118024B2 (en) | 2024-10-15 |
US20240184815A1 (en) | 2024-06-06 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN112313697B (zh) | 用于生成描述角度增强的可解释的基于描述的推荐的系统和方法 | |
US10824942B1 (en) | Visual similarity and attribute manipulation using deep neural networks | |
US12111850B2 (en) | Genealogical entity resolution system and method | |
KR102119253B1 (ko) | 영상데이터의 추상적특성 획득 방법, 장치 및 프로그램 | |
EP2895967A1 (en) | Gesture-based search queries | |
US11210327B2 (en) | Syntactic profiling of alphanumeric strings | |
WO2020255307A1 (ja) | 情報処理装置、情報処理方法、および記録媒体 | |
WO2010026900A1 (ja) | 関係性発見装置、関係性発見方法および記録媒体 | |
TWI665630B (zh) | 互動式商品推薦方法以及非暫態電腦可讀取媒體 | |
US10417798B2 (en) | System and method based on sliding-scale cluster groups for precise look-alike modeling | |
WO2022201541A1 (ja) | 検索装置、検索方法、及びコンピュータ読み取り可能な記録媒体 | |
JP7451157B2 (ja) | 情報処理装置、情報処理方法、及びプログラム | |
CN109062970A (zh) | 用户画像的生成方法、生成设备及计算机可读存储介质 | |
US20210271637A1 (en) | Creating descriptors for business analytics applications | |
JP5559750B2 (ja) | 広告処理装置、情報処理システム及び広告処理方法 | |
CN109299378B (zh) | 搜索结果的展示方法、装置、终端及存储介质 | |
KR20200141387A (ko) | 딥러닝 알고리즘을 이용한 영상데이터 검색방법, 장치 및 프로그램 | |
WO2020231091A1 (en) | Improved exploration for interactive recommendation systems | |
CN114020245A (zh) | 页面构建方法及装置、设备和介质 | |
CN112051951A (zh) | 一种媒体内容展示方法、媒体内容的展示确定方法及装置 | |
JP4871650B2 (ja) | アイテムデータを送信する方法、サーバおよびプログラム | |
JP2017004260A (ja) | 情報処理装置、情報処理方法及び情報処理プログラム | |
Çiçek et al. | Deterioration of pre-war and rehabilitation of post-war urbanscapes using generative adversarial networks | |
CN111797257B (zh) | 基于词向量的图片推荐方法及相关设备 | |
US20240346233A1 (en) | Efficient generation of review summaries |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 21933142 Country of ref document: EP Kind code of ref document: A1 |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2023508410 Country of ref document: JP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 18278762 Country of ref document: US |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 21933142 Country of ref document: EP Kind code of ref document: A1 |