Keyword extraction-based AI intelligent question-answering system and method

Abstract

The invention relates to the field of intelligent question answering, in particular to an AI intelligent question answering system and method based on keyword extraction. Comprising the steps of obtaining an input statement; analyzing the structure of the statement, and extracting a plurality of keywords associated with each other; classifying according to semantic elements of the keywords, and screening out three keywords with the highest actual meaning proportion; and expanding a standard question based on the three extraction words, and inputting the standard question into the corresponding question and answer model training to obtain a result. Through intelligent keyword extraction and screening, the understanding accuracy of the question and answer system on the input statement can be improved, and it is ensured that the extracted cue word accurately reflects the core demand of the user.

Description

Technical Field

[0001] The present invention relates to the field of intelligent question answering technology, and in particular to an AI intelligent question answering system and method based on keyword extraction. Background Art

[0002] With the rapid development of artificial intelligence (AI) technology, the application of intelligent question-answering systems in various fields has garnered widespread attention. In particular, in areas such as information query, customer service, online education, and intelligent assistants, intelligent question-answering systems have become an essential tool for problem-solving and service delivery. However, current intelligent question-answering systems still face several technical challenges, particularly in accurately extracting key information from massive amounts of information to provide high-quality answers. Numerous bottlenecks remain.

[0003] Existing intelligent question-answering systems mostly rely on keyword matching technology or simple rule-based systems. These traditional methods perform well for standard questions, but they often fall short for complex, open-ended questions. In particular, when questions contain ambiguity, synonyms, or are highly context-dependent, rule-based systems struggle to accurately extract useful information, significantly reducing the accuracy and intelligence of the answers.

[0004] In recent years, with the advancement of natural language processing (NLP) technology, deep learning-based question-answering systems have become a research hotspot. By introducing pre-trained models (such as BERT and GPT), question-answering systems can achieve a certain degree of semantic understanding and contextual reasoning. However, these methods typically require extensive computing resources and complex model training, which still poses significant challenges in resource-constrained scenarios or when real-time response is required.

[0005] Therefore, the key to improving the performance of intelligent question-answering systems is how to combine semantic analysis and contextual information to accurately extract keywords while ensuring high efficiency and low computational cost. Based on this, this paper proposes an AI intelligent question-answering system and method based on keyword extraction. This aims to improve the intelligence level and response efficiency of the question-answering system through an innovative keyword extraction algorithm, thereby addressing the shortcomings of existing technologies. Summary of the Invention

[0006] The present invention provides an AI-powered intelligent question-answering system and method based on keyword extraction, aiming to address the existing shortcomings of intelligent question-answering systems, which often prevent users from accurately asking questions. Through intelligent keyword extraction and standard question expansion, this application improves the accuracy and intelligence of the question-answering system, optimizes information classification and filtering, and enhances the system's contextual understanding capabilities, while also improving efficiency and reducing computational costs.

[0007] This application provides an AI intelligent question-answering method based on keyword extraction, including: Get the input statement; Analyzing the structure of the sentence and extracting multiple keywords related to each other; Classify keywords based on their semantic elements and select three related keywords; Based on the three extracted words, a standard question is expanded, and the standard question is input into the corresponding question-answering model training to obtain the result.

[0008] The analyzing the structure of the sentence and extracting multiple related keywords include: Preprocessing the statement to remove useless information; Obtaining the cohesion of the statement after preprocessing; Information entropy is obtained through the cohesion to determine multiple keywords.

[0009] According to the semantic elements of the keywords, three related keywords were selected, including: A quick lookup table of keyword and basic element classification is preset; Based on the quick lookup table of keyword basic element classification, three interrelated keywords were selected.

[0010] The standard question is expanded based on the three extracted words, and the standard question is input into the corresponding question-answering model training, and the results include: An element combination table is preset, and a standard question sentence is expanded based on the element combination table. The standard question sentence is input into the question-answering model to obtain the question-answering result.

[0011] It also includes a semantic adjustment model for semantically adjusting the question and answer results to obtain a standard answer.

[0012] Another object of the present invention is to provide an AI intelligent question-answering system based on keyword extraction, which is applied to the above-mentioned AI intelligent question-answering method based on keyword extraction, comprising: The acquisition module is used to obtain the input statement; An extraction module, configured to analyze the structure of the input sentence and extract keywords from the input sentence; A classification module, used to classify the keywords and select three mutually related keywords; The question-answering module is used to expand the three interrelated keywords into a standard sentence and input the sentence into the question-answering model to obtain a result.

[0013] The extraction module includes: Preprocessing module: used to preprocess input sentences and remove useless information; Calculation module: obtains the cohesion of the input sentence after preprocessing; Keyword module: obtain information entropy through cohesion to determine multiple keywords; The classification module includes: The screening module selects three interrelated classification words through a preset keyword basic element classification quick lookup table.

[0014] The question-answering module includes: An adjustment module, used for expanding three interrelated keywords into a standard question sentence based on an element combination table; The input module is used to input standard questions into the question-answering model to obtain question-answering results.

[0015] It also includes a semantic adjustment module for semantically adjusting the question and answer results to obtain standard answers.

[0016] Compared with the prior art, the AI intelligent question-answering system and method based on keyword extraction provided by the present invention has the following beneficial effects: Through intelligent keyword extraction and screening, the question-answering system can improve its understanding of input sentences, ensuring that the extracted prompt words accurately reflect the user's core needs. Secondly, by generating standardized question statements based on the element combination table, the question-answering model can more efficiently handle different forms of questions, improving the system's flexibility and adaptability. By optimizing the semantic adjustment model, it can ensure that the question-answering results are more consistent with user intent, improving the accuracy and relevance of the answers. In addition, this method reduces dependence on complex computing resources, improves computing efficiency, and supports real-time response and dynamic learning, thereby further improving the user experience and the overall performance of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 This is a flow chart of the AI intelligent question-answering method based on keyword extraction of the present invention; Figure 2 This is a structural block diagram of the AI intelligent question-answering system based on keyword extraction of the present invention. DETAILED DESCRIPTION

[0018] The following embodiments of the present invention are described in further detail with reference to the accompanying drawings and examples. The following examples are used to illustrate the present invention but are not intended to limit the scope of the present invention.

[0019] In the description of this application, it should be understood that the terms "center", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating the orientation or position relationship, are based on the orientation or position relationship shown in the accompanying drawings, and are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on this application.

[0020] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of the technical features being referred to. Thus, a feature specified as "first" or "second" may explicitly or implicitly include one or more of such features. Throughout this application, unless otherwise specified, "plurality" means two or more.

[0021] In the description of this application, it should be noted that, unless otherwise expressly specified or limited, the terms "mounted," "connected," and "connected" should be understood in a broad sense. For example, they can refer to fixed connections, detachable connections, or integral connections; mechanical connections or electrical connections; direct connections or indirect connections through an intermediate medium; and internal connections between two components. Those skilled in the art will understand the specific meanings of the above terms in this application based on the specific circumstances.

[0022] like Figure 1-Figure 2 As shown, the present invention provides an AI intelligent question-answering system and method based on keyword extraction, which is used to solve the shortcoming in the prior art that people cannot accurately ask questions when using intelligent question-answering systems.

[0023] The present invention provides an AI intelligent question-answering method based on keyword extraction, comprising: Get the input statement; Analyzing the structure of the sentence and extracting multiple keywords related to each other; Classify keywords based on their semantic elements and select three related keywords; Based on the three keywords, standard questions are expanded and input into the corresponding question-answering model training to obtain the results.

[0024] The analyzing the structure of the sentence and extracting multiple related keywords include: Preprocessing the statement to remove useless information; Obtaining the cohesion of the statement after preprocessing; Information entropy is obtained through the cohesion to determine multiple keywords.

[0025] According to the semantic elements of the keywords, three related keywords were selected, including: A quick lookup table of keyword basic element classification is preset; the quick lookup table of basic element classification is shown in Table 1.

[0026] Table 1 Based on the quick lookup table of keyword basic element classification, three interrelated keywords are selected.

[0027] The standard questions are expanded based on keywords and input into the corresponding question-answering model training, and the results include: An element combination table is preset. Based on the element combination table, the three keywords are expanded into standard questions, and the standard questions are input into the question-answering model to obtain the question-answering results.

[0028] The element combination table is shown in Table 2.

[0029] Table 2 It also includes a semantic adjustment model for semantically adjusting the question and answer results to obtain a standard answer.

[0030] The present invention also provides an AI intelligent question-answering system based on keyword extraction, which is applied to the above-mentioned AI intelligent question-answering method based on keyword extraction, comprising: The acquisition module is used to obtain the input statement; An extraction module, configured to analyze the structure of the input sentence and extract keywords from the input sentence; A classification module, used to classify the keywords and select three mutually related keywords; The question-answering module is used to expand the three interrelated keywords into a standard sentence and input the sentence into the question-answering model to obtain a result.

[0031] The extraction module includes: Preprocessing module: used to preprocess input sentences and remove useless information; Calculation module: obtains the cohesion of the input sentence after preprocessing; Keyword module: Get information entropy through cohesion to determine multiple keywords The classification module includes: The screening module selects three interrelated classification words through a preset keyword basic element classification quick lookup table.

[0032] The question-answering module includes: An adjustment module, used for expanding three interrelated keywords into a standard question sentence based on an element combination table; The input module is used to input standard questions into the question-answering model to obtain question-answering results.

[0033] It also includes a semantic adjustment module for semantically adjusting the question and answer results to obtain standard answers.

[0034] Example 1 The sentence question entered by the user is "I am currently preparing for the college entrance examination, and I particularly hope to improve my math scores. In order to improve my math level in a more organized way, I plan to make a detailed study plan. I hope this plan can be carried out on a weekly basis, and it will specify what to do each week and how to effectively improve my math ability. I hope there will be some practical methods and steps to help me gradually make up for my shortcomings in math and improve my problem-solving ability."

[0035] The system extracts multiple interrelated prompt words from the input sentence according to the sentence structure, classifies the prompt words according to the elements, and selects the three prompt words with the highest actual meaning weight: "College Entrance Examination Mathematics", "Weekly", and "Improvement Methods". Then, based on the element combination table, it expands the most standard question sentence composed of the three selected prompt words, and inputs the question sentence as a training sample into the corresponding question-answering training model. The result output by the question-answering training model is adjusted by the semantic adjustment model to obtain the standard answer to the input sentence.

[0036] The final standard question output is "I am a student preparing to take the college entrance examination. I want to improve my math scores. Please list specific study plans and methods in points on a weekly basis."

[0037] Example 2 The sentence question entered by the user is "I hope you can design a complete menu for a newly opened restaurant from the perspective of a senior planner. This menu needs to be simple and beautiful, highlighting the restaurant's special dishes, and each dish should include a detailed introduction to the ingredients and prices. In order to make this menu look modern and attractive, you can refer to the menu style of restaurants like Haidilao, ensuring that it is visually simple and attractive, while clearly conveying the core characteristics and uniqueness of each dish. Ultimately, the goal is to create a menu that is easy for customers to choose and perfectly showcases the restaurant's personality."

[0038] The system extracts multiple interrelated prompt words from the input sentence based on the sentence structure, classifies the prompt words according to the elements, and screens out the three prompt words with the highest actual meaning weights. Then, based on the element combination table, it expands the most standard question sentence composed of the three screened prompt words, inputs the question sentence as a training sample into the corresponding question-answering training model, and adjusts the result output by the question-answering training model through the semantic adjustment model to obtain the standard answer to the input sentence.

[0039] The final standard question and answer is "Please design a complete menu for a newly opened restaurant from the perspective of a senior planner. The menu should be simple and beautiful, highlight the characteristics of the dishes, and include prices and ingredient introductions. You can refer to Haidilao's menu style."

[0040] Example 3 The sentence question entered by the user is "Please consider recommending several suitable laptops to college students. Because college students have many usage needs, not only do they need to meet basic needs such as study and office work, but they also need to involve light gaming and entertainment or some design and programming tasks. So when recommending a laptop, I hope you will consider the three aspects of performance, portability, and price to ensure that the recommended computer can meet the needs of college students in these aspects."

[0041] The system extracts multiple interrelated prompt words from the input sentence based on the sentence structure, classifies the prompt words according to the elements, and screens out the three prompt words with the highest actual meaning weights. Then, based on the element combination table, it expands the most standard question sentence composed of the three screened prompt words, inputs the question sentence as a training sample into the corresponding question-answering training model, and adjusts the result output by the question-answering training model through the semantic adjustment model to obtain the standard answer to the input sentence.

[0042] The final standard question and answer is "Please recommend several laptops suitable for college students. When recommending, we hope you can explain in detail the advantages and disadvantages of each computer in terms of performance, portability, and price. If the recommended computer does not meet the mainstream usage scenarios of college students such as gaming, study, office, etc., please recommend it again and explain the reasons."

[0043] The above description is only an embodiment of the present invention, but it cannot be used to limit the scope of the present invention. Any structural changes made according to the present invention should be deemed to fall within the scope of protection of the present invention and be subject to restrictions as long as they do not lose the essence of the present invention.

[0044] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working process and related instructions of the system described above can refer to the corresponding process in the aforementioned method embodiment and will not be repeated here.

[0045] It should be noted that the system provided in the above embodiment is only illustrated by the division of the above functional modules. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the modules or steps in the embodiments of the present invention can be further decomposed or combined. For example, the modules in the above embodiment can be combined into one module, or further divided into multiple sub-modules to complete all or part of the functions described above. The names of the modules and steps involved in the embodiments of the present invention are only for distinguishing the modules or steps and are not to be regarded as improper limitations of the present invention.

[0046] Those skilled in the art should be able to appreciate that the modules and method steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two, and the programs corresponding to the software modules and method steps can be placed in random access memory (RAM), internal memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disks, removable disks, CD-ROMs, or any other form of storage medium known in the art. In order to clearly illustrate the interchangeability of electronic hardware and software, the above description has generally described the components and steps of each example according to their functions. Whether these functions are performed in electronic hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art may use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0047] The term "comprise" or any other similar term is intended to cover non-exclusive inclusion such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed or inherent to such process, method, article, or apparatus.

[0048] Thus far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art may make equivalent changes or substitutions to the relevant technical features, and the technical solutions after such changes or substitutions will fall within the scope of protection of the present invention.

[0049] The above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention.

Claims

1. An AI intelligent question answering method based on keyword extraction, characterized in that: include: Get the input statement; Analyzing the structure of the sentence and extracting multiple keywords related to each other; Classify keywords based on their semantic elements and select three related keywords; Based on the three keywords, standard questions are expanded and input into the corresponding question-answering model training to obtain the results.

2. The AI intelligent question-answering method based on keyword extraction according to claim 1, characterized in that: The analyzing the structure of the sentence and extracting multiple related keywords include: Preprocessing the statement to remove useless information; Obtaining the cohesion of the statement after preprocessing; Information entropy is obtained through the cohesion to determine multiple keywords.

3. The AI intelligent question-answering method based on keyword extraction according to claim 1, characterized in that: According to the semantic elements of the keywords, three related keywords were selected, including: A quick lookup table of keyword and basic element classification is preset; Based on the quick lookup table of keyword basic element classification, three interrelated keywords are selected.

4. The AI intelligent question-answering method based on keyword extraction according to claim 1, characterized in that: The standard questions are expanded based on keywords and input into the corresponding question-answering model training, and the results include: An element combination table is preset. Based on the element combination table, the three keywords are expanded into standard questions, and the standard questions are input into the question-answering model to obtain the question-answering results.

5. The AI intelligent question-answering method based on keyword extraction according to claim 1, characterized in that: It also includes a semantic adjustment model for semantically adjusting the question and answer results to obtain a standard answer.

6. An AI intelligent question-answering system based on keyword extraction, applied to the AI intelligent question-answering method based on keyword extraction according to any one of claims 1 to 5, characterized in that: include: The acquisition module is used to obtain the input statement; An extraction module, configured to analyze the structure of the input sentence and extract keywords from the input sentence; A classification module, used to classify the keywords and select three mutually related keywords; The question-answering module is used to expand the three interrelated keywords into a standard sentence and input the sentence into the question-answering model to obtain a result.

7. The AI intelligent question-answering system based on keyword extraction according to claim 6, characterized in that: The extraction module includes: Preprocessing module: used to preprocess input sentences and remove useless information; Calculation module: obtains the cohesion of the input sentence after preprocessing; Keyword module: The information entropy is obtained through cohesion to determine multiple keywords.

8. The AI intelligent question-answering system based on keyword extraction according to claim 6, characterized in that: The classification module includes: The screening module selects three interrelated classification words through a preset keyword basic element classification quick lookup table.

9. The AI intelligent question-answering system based on keyword extraction according to claim 6, characterized in that: The question-answering module includes: An adjustment module, used for expanding three interrelated keywords into a standard question sentence based on an element combination table; The input module is used to input standard questions into the question-answering model to obtain question-answering results.

10. The AI intelligent question-answering system based on keyword extraction according to claim 6, characterized in that: It also includes a semantic adjustment module for semantically adjusting the question and answer results to obtain standard answers.

Images