Information processing device

Abstract

If an inappropriate answer is generated, the answer will be automatically regenerated. [Solution] The information processing device includes: a first extraction means for extracting a first summary feature, which is a feature of the summary of the question entered by the user, and a first keyword feature, which is a feature of the keywords included in the question; a second extraction means for extracting a second summary feature, which is a feature of the summary of the answer generated by the model in response to the question, and a second keyword feature, which is a feature of the keywords included in the answer; a determination means for determining whether the distance between the first summary feature and the first keyword feature and the second summary feature and the second keyword feature exceeds a predetermined threshold; and an instruction means for instructing the model to regenerate the answer if the distance exceeds a predetermined threshold.

Description

Technical Field

[0001] This disclosure relates to the technical field of information processing apparatuses.

Background Art

[0002] As this type of apparatus, there is known one that outputs an answer to a question from a user. For example, Patent Document 1 discloses a technique for automatically analyzing a question email from a mobile phone and automatically creating and transmitting a response sentence based on the analysis result.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] When generating an answer using a machine-learned model for a question from a user, an inappropriate answer may be generated. In this case, there arise technical problems such as multiple exchanges (for example, re-input of the question sentence, etc.) between the user and the model, and an increase in the time and effort required to obtain an appropriate answer.

[0005] This disclosure has been made in view of the above problems, and an object thereof is to provide an information processing apparatus capable of automatically regenerating an answer when an inappropriate answer is generated.

Means for Solving the Problems

[0006] An information processing device according to one aspect of this disclosure includes: a first extraction means for extracting a first summary feature, which is a feature of the summary of a question entered by a user, and a first keyword feature, which is a feature of the keywords contained in the question; a second extraction means for extracting a second summary feature, which is a feature of the summary of the answer generated by a model for the question, and a second keyword feature, which is a feature of the keywords contained in the answer; a determination means for determining whether the distance between the first summary feature and the first keyword feature and the second summary feature and the second keyword feature exceeds a predetermined threshold; and an instruction means for instructing the model to regenerate the answer if the distance exceeds a predetermined threshold. [Brief explanation of the drawing]

[0007] [Figure 1] This is a block diagram showing the hardware configuration of an information processing device according to an embodiment. [Figure 2] This is a block diagram showing the functional configuration of an information processing apparatus according to an embodiment. [Figure 3] This is a flowchart showing the operation flow of the information processing device according to the embodiment. [Modes for carrying out the invention]

[0008] The following describes an embodiment of the information processing device with reference to the drawings.

[0009] (Hardware configuration) First, the hardware configuration of the information processing device according to the embodiment will be described with reference to Figure 1. Figure 1 is a block diagram showing the hardware configuration of the information processing device according to the embodiment.

[0010] In Figure 1, the information processing device 10 according to this embodiment comprises an arithmetic unit 110, a storage device 120, a communication device 130, an input device 140, and an output device 150. The arithmetic unit 110, the storage device 120, the communication device 130, the input device 140, and the output device 150 are connected to each other via a data bus.

[0011] The arithmetic unit 110 is configured to perform various arithmetic operations in the information processing device 10. The arithmetic unit 110 may have a processor. The arithmetic unit 110 may have a single processor or may have multiple processors. In other words, the arithmetic unit 110 may have one or more processors. The processor may be a multi-core processor. If the arithmetic unit 110 has a single processor that is a multi-core processor, then logically, the arithmetic unit 110 can be said to have multiple processors.

[0012] The processor in the arithmetic unit 110 may be at least one of the following: CPU (Central Processing Unit), GPU (Graphics Processing Unit), FPGA (Field Programmable Gate Array), and TPU (Tensor Processing Unit).

[0013] The storage device 120 may be at least one of the following: RAM (Random Access Memory), ROM (Read Only Memory), hard disk drive, magneto-optical disk drive, SSD (Solid State Drive), and optical disk array. In other words, the storage device 120 may be implemented by a single device or by multiple devices.

[0014] The storage device 120 is capable of storing desired data. The storage device 120 may store the computer program CP that the arithmetic unit 110 will execute. The storage device 120 may temporarily store data that the arithmetic unit 110 will use temporarily when the arithmetic unit 110 is executing the computer program CP.

[0015] The computer program CP may be recorded on a non-temporary recording medium that is readable by a computer. In this case, the computer program CP may be stored in the storage device 120 by reading the recording medium using a recording medium reading device (not shown) provided by the information processing device 10. At least one of the following may be used as the recording medium: an optical disc, a magnetic medium, a magneto-optical disc, a semiconductor memory, and any other medium capable of storing a program. The computer program CP may also be obtained from an external device (not shown) of the information processing device 10 via a communication device 130. In other words, the computer program CP may be downloaded from an external device to the storage device 120 of the information processing device 10.

[0016] The arithmetic unit 110 (for example, a processor) may execute the processing that the information processing device 10 should perform together with the memory device 120 in which the computer program CP is stored (in other words, together with the memory device 120 and the computer program CP stored in the memory device 120). For example, by the arithmetic unit 110 executing the computer program CP, a logical functional block for executing the processing that the information processing device 10 should perform may be realized within the arithmetic unit 110 (for example, within the processor).

[0017] The communication device 130 is configured to communicate with devices outside the information processing device 10. The communication device 130 may use wired communication or wireless communication.

[0018] The input device 140 is a device capable of receiving input of information to the information processing device 10 from the outside. The input device 140 may include an operating device (for example, a keyboard, a mouse, a touch panel, etc.) that can be operated by a user of the information processing device 10. The input device 140 may include a recording medium reader that can read information recorded on a detachable recording medium of the information processing device 10, such as a USB (Universal Serial Bus) memory. When information is input to the information processing device 10 via the communication device 130 (in other words, when the information processing device 10 acquires information via the communication device 130), the communication device 130 may function as an input device.

[0019] The output device 150 is a device capable of outputting information to the outside of the information processing device 10. The output device 150 may have a display device capable of outputting visual information such as characters and images as the above information. The output device 150 may also have a speaker capable of outputting auditory information such as voices as the above information. The output device 150 may be configured to output the above information (for example, control information of other devices, etc.) to other devices. The output device 150 may be capable of outputting information to a detachable recording medium of the information processing device 10, such as a USB memory. When the information processing device 10 outputs information via the communication device 130, the communication device 130 may function as an output device.

[0020] <Functional Configuration> Next, the functional configuration of the information processing device 10 according to the embodiment will be described while referring to FIG. 2. FIG. 2 is a block diagram showing the functional configuration of the information processing device according to the embodiment.

[0021] In FIG. 2, the information processing apparatus 10 is configured as a device that instructs the regeneration of a response sentence when the response sentence generated for the question sentence input by the user is inappropriate. The information processing apparatus 10 includes, as components for realizing its functions, a question sentence acquisition unit 210, a first extraction unit 220, a question sentence output unit 230, a response sentence acquisition unit 240, a second extraction unit 250, a determination unit 260, a regeneration instruction unit 270, and a response sentence output unit 280. Note that each of the question sentence acquisition unit 210, the first extraction unit 220, the question sentence output unit 230, the response sentence acquisition unit 240, the second extraction unit 250, the determination unit 260, the regeneration instruction unit 270, and the response sentence output unit 280 may be a processing block realized by the arithmetic unit 110 described above.

[0022] The question sentence acquisition unit 210 is configured to be able to acquire the question sentence input by the user. For example, the question sentence acquisition unit 210 may acquire the text input by the user using a keyboard or the like as the question sentence. Alternatively, the question sentence acquisition unit 210 may acquire the voice input by the user using a microphone or the like as the question sentence. In this case, the question sentence acquisition unit 210 may have a voice recognition function for converting voice data into text data. The question sentence acquired by the question sentence acquisition unit 210 is output to the first extraction unit 220 and the question sentence output unit 230.

[0023] The first extraction unit 220 is configured to be able to extract a first summary feature amount and a first keyword feature amount from the question sentence acquired by the question sentence acquisition unit 210. The first summary feature amount is a feature amount of the summary of the question sentence. The first keyword feature amount is a feature amount of the keyword included in the question sentence. The first summary feature amount and the first keyword feature amount may be extracted as feature vectors indicating features. Note that the method for extracting the first summary feature amount and the first keyword feature amount is not particularly limited. The first extraction unit 220 may extract the first summary feature amount and the first keyword feature amount using, for example, a machine-learned model. Each of the first summary feature amount and the first keyword feature amount extracted by the first extraction unit 220 is output to the determination unit 260.

[0024] The question output unit 230 is configured to output the question text acquired by the question acquisition unit 210 to the answer generation model 300. The answer generation model 300 is a model that generates an answer text for a question text. The answer generation model may be, for example, a Large Language Model (LLM). Although Figure 2 shows an example where the answer generation model 300 is located outside the information processing unit 10, the information processing unit 10 may also be configured to include the answer generation model 300.

[0025] The response text acquisition unit 240 is configured to acquire the response text generated by the response generation model 300. The response text acquired by the response text acquisition unit 240 is output to the second extraction unit 260. The response text acquisition unit 240 is also configured to output the response text to the response text output unit 230 based on the determination result in the determination unit 260.

[0026] The second extraction unit 250 is configured to extract a second summary feature and a second keyword feature from the response text acquired by the response text acquisition unit 240. The second summary feature is a feature of the summary of the response text. The second keyword feature is a feature of the keywords included in the response text. The second summary feature and the second keyword feature may be extracted as feature vectors that represent the features. The method for extracting the second summary feature and the second keyword feature is not particularly limited. The second extraction unit 250 may, for example, use a machine learning model to extract the second summary feature and the second keyword feature. Each of the second summary feature and the second keyword feature extracted by the second extraction unit 250 is configured to be output to the determination unit 260.

[0027] The determination unit 260 is configured to calculate the distance between the first summary feature and the first keyword feature and the second summary feature and the second keyword feature. The distance here may be a value indicating the similarity or relevance between the first summary feature and the first keyword feature and the second summary feature and the second keyword feature. In this case, the larger the calculated distance, the lower the similarity or relevance between the first summary feature and the first keyword feature and the second summary feature and the second keyword feature. The determination unit 260 may calculate the distance between the first summary feature and the second summary feature and the distance between the first keyword feature and the second keyword feature separately. Alternatively, the determination unit 260 may calculate the distance between the first summary feature and the first keyword feature and the second summary feature and the second keyword feature together. For example, the determination unit 260 may calculate the distance between the feature obtained by integrating the first summary feature and the first keyword feature and the feature obtained by integrating the second summary feature and the second keyword feature.

[0028] The determination unit 260 is further configured to determine whether the calculated distance exceeds a predetermined threshold. The predetermined threshold here is a threshold for determining whether the answer sentence generated by the answer generation model is appropriate for the question sentence entered by the user. For example, if the distance between the first summary feature and the first keyword feature and the second summary feature and the second keyword feature exceeds the predetermined threshold, it can be determined that the similarity or relevance between the question sentence and the answer sentence is low (i.e., the answer sentence is not appropriate). On the other hand, if the distance between the first summary feature and the first keyword feature and the second summary feature and the second keyword feature does not exceed the predetermined threshold, it can be determined that the similarity or relevance between the question sentence and the answer sentence is high (i.e., the answer sentence is appropriate). The determination result from the determination unit 260 is output to the answer sentence acquisition unit 240 and the regeneration instruction unit 270.

[0029] The regeneration instruction unit 270 is configured to output an instruction to the answer generation model 300 to regenerate the answer text. Specifically, the regeneration instruction unit 270 outputs an instruction to regenerate the answer text when the distance calculated by the determination unit 260 exceeds a predetermined threshold. The regeneration instruction unit 270 may, for example, instruct the answer generation model 300 to regenerate the answer text by modifying a part of the question text entered by the user and outputting it to the answer generation model 300.

[0030] The response output unit 280 is configured to output a response to the user who entered the question. Specifically, the response output unit 280 outputs a response to the user if the distance calculated by the determination unit 260 does not exceed a predetermined threshold.

[0031] (Flow of operations) Next, the operation flow of the information processing device 10 according to the embodiment will be described with reference to Figure 3. Figure 3 is a flowchart showing the operation flow of the information processing device according to the embodiment.

[0032] As shown in Figure 3, when the operation of the information processing device 10 according to the embodiment is started, first the question text acquisition unit 210 acquires the question text entered by the user (step S101). Then, the first extraction unit 220 extracts a first summary feature and a first keyword feature from the question text acquired by the question text acquisition unit 210 (step S102). In addition, the question text output unit 230 outputs the question text acquired by the question text acquisition unit 210 to the answer generation model 300 (step S103). As a result, the answer generation model 300 generates an answer text for the question text.

[0033] Next, the response text acquisition unit 240 acquires the response text generated by the response generation model 300 (step S104). Then, the second extraction unit 250 extracts the second summary feature and the second keyword feature from the response text acquired by the response text acquisition unit 240 (step S105).

[0034] Next, the determination unit 260 calculates the distance between the first summary feature and the first keyword feature and the second summary feature and the second keyword feature, and determines whether the calculated distance exceeds a predetermined threshold (step S106). If the distance calculated by the determination unit 260 does not exceed the predetermined threshold (step S106: NO), the response output unit 280 outputs the response obtained by the response acquisition unit 240 to the user (step S107). In this case, regeneration of the response is not required.

[0035] On the other hand, if the distance calculated by the determination unit 260 exceeds a predetermined threshold (step S106: YES), the regeneration instruction unit 270 outputs an instruction to the answer generation model 300 to regenerate the answer (step S108). As a result, the answer generation model 300 generates a new answer sentence for the question sentence.

[0036] If the response is regenerated by the response generation model 300, the process from step S104 will be repeated. After repeating this process, if the distance calculated by the determination unit 260 no longer exceeds a predetermined threshold (step S106: NO), the response output unit 280 outputs the response acquired by the response acquisition unit 240 (i.e., the regenerated response) to the user (step S107).

[0037] (Technical effects) Next, the technical effects obtained by the information processing device 10 according to this embodiment will be described.

[0038] As explained in Figures 1 to 3, the information processing device 10 according to the embodiment determines whether the distance between the first summary feature and first keyword feature extracted from the question and the second summary feature and second keyword feature extracted from the answer exceeds a predetermined distance. This makes it possible to automatically determine whether the answer is appropriate to the question. Furthermore, in the information processing device 10 according to the embodiment, if the distance calculated by the determination unit 260 exceeds a predetermined distance, an instruction is output to the answer generation model 300 to reproduce the answer. As a result, even if the answer generated by the answer generation model 300 is inappropriate, the answer is repeatedly regenerated until it becomes appropriate. As a result, it becomes possible to output an appropriate answer to the user.

[0039] Furthermore, if it were not possible to automatically determine whether the answer is appropriate or not, the user would be required to, for example, revise and re-enter the question. In this case, the user's effort and the time it takes to obtain an appropriate answer would increase, and as a result, the user may lose trust in the answer generation model 300. However, in this embodiment, as described above, it is possible to automatically determine whether the answer is appropriate or not. Therefore, it is possible to output an appropriate answer to the user while avoiding the aforementioned inconveniences.

[0040] The embodiments of the invention derived from the above-described embodiments are described below.

[0041] An information processing device according to one aspect of this disclosure includes: a first extraction means for extracting a first summary feature, which is a feature of the summary of a question input by a user, and a first keyword feature, which is a feature of the keywords contained in the question; a second extraction means for extracting a second summary feature, which is a feature of the summary of the answer generated by a model for the question, and a second keyword feature, which is a feature of the keywords contained in the answer; a determination means for determining whether the distance between the first summary feature and the first keyword feature and the second summary feature and the second keyword feature exceeds a predetermined threshold; and an instruction means for instructing the model to regenerate the answer if the distance exceeds a predetermined threshold. In the above embodiment, the "first extraction unit 220" corresponds to an example of the "first extraction means," the "second extraction unit 250" corresponds to an example of the "second extraction means," the "determination unit 260" corresponds to an example of the "determination means," and the "regeneration instruction unit 270" corresponds to an example of the "instruction means."

[0042] In the information processing device according to the above embodiment, the device may further include output means for outputting the regenerated answer sentence to the user when the distance between the first summary feature and the first keyword feature and the second summary feature and the second keyword feature extracted from the regenerated answer sentence is less than or equal to a predetermined threshold. In this way, when an appropriate answer sentence is generated (for example, when the answer sentence becomes appropriate through regeneration), that answer sentence can be output to the user. In the above embodiment, the "answer sentence output unit 280" corresponds to an example of the "output means".

[0043] This disclosure is not limited to the embodiments described above and can be modified as appropriate without contradicting the gist or idea of ​​the invention as can be inferred from the claims and the specification as a whole. Information processing devices with such modifications are also included within the technical scope of the present invention. [Explanation of symbols]

[0044] 10...Information processing device, 110...Calculation unit, 120...Storage device, 130...Communication device, 140...Input device, 150...Output device, 210...Question text acquisition unit, 220...First extraction unit, 230...Question text output unit, 240...Answer text acquisition unit, 250...Second extraction unit, 260...Determination unit, 270...Regeneration instruction unit, 280...Answer text output unit, 300...Answer generation model

Claims

1. A first extraction means for extracting a first summary feature, which is a feature of the summary of the question entered by the user, and a first keyword feature, which is a feature of the keywords included in the question. A second extraction means for extracting a second summary feature, which is a feature of the summary of the answer sentence generated by the model in response to the aforementioned question sentence, and a second keyword feature, which is a feature of the keywords contained in the answer sentence. A determination means for determining whether the distance between the first summary feature and the first keyword feature and the second summary feature and the second keyword feature exceeds a predetermined threshold, An instruction means for instructing the model to regenerate the response sentence when the distance exceeds a predetermined threshold, An information processing device equipped with the following features.

2. The system further includes output means for outputting the regenerated response text to the user when the distance between the first summary feature and the first keyword feature and the second summary feature and the second keyword feature extracted from the regenerated response text is less than or equal to a predetermined threshold. The information processing apparatus according to claim 1.

Images