Method and device for automatically designing tatami according to house type
Through the generation and scoring of tatami design solutions through multi-layer artificial intelligence systems, the design consuming and labor-intensive and error-prone problems in the existing technology are solved, and the rapid and accurate design solutions are achieved, which improves the user experience.
Patent Information
- Application Number
- CN202411152918.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-21
- Publication Date
- 2024-11-15
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Existing design software requires a lot of time and manual intervention when designing tatami, which is prone to errors and is difficult to quickly provide the optimal solution, which cannot meet the rapid needs of owners.
Using a multi-layer artificial intelligence system, multiple tatami design solutions are generated by obtaining house types and user needs, and using the second AI to score the solutions and select solutions with a score higher than the predetermined threshold, overcoming the time-consuming and labor-intensive design and easy to make mistakes. flaws.
It realizes the fast and accurate provision of user-satisfied tatami design solutions, improves user experience, and reduces design time and error rate.
Smart Images

Figure CN118965539A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field related to home customization, and more specifically, to a method and device for automatically designing tatami according to the house type. Background Art
[0002] At present, the existing design software for tatami design requires the long-term work experience of the custom designer to determine how to layout the tatami in a bedroom. Sometimes various layouts need to be tried repeatedly, and the designer needs to manually calculate the size of each area and the remaining aisle size. This not only takes a lot of time, but also the resulting design plan is not necessarily the best plan, and it is easy to make mistakes, resulting in problems such as insufficient aisle size. It is impossible to provide the owner with a quick tatami bedroom plan in a short time. Therefore, it is necessary to design a technical solution that can overcome the above defects. Summary of the invention
[0003] An object of the present invention is to provide a method and device for automatically designing tatami according to the house type, which can overcome the defects of the prior art tatami design that is time-consuming, labor-intensive and prone to errors, and improve the user experience.
[0004] In order to achieve these purposes and other advantages of the present invention, according to one aspect of the present invention, the present invention provides a method for automatically designing tatami according to a house type, comprising: S1: obtaining house type parameters and user requirements of a target house type; S2: using a first AI to generate a plurality of tatami design schemes according to the house type parameters and the user requirements; S3: using a second AI (artificial intelligence) to score the plurality of tatami design schemes; S4: selecting the tatami design scheme having a score higher than a predetermined threshold.
[0005] Furthermore, the apartment parameters include apartment size and door and window positions, and the user requirements include design style and design color.
[0006] Furthermore, the first AI is trained through samples of multiple apartment parameters and user requirements and corresponding tatami design solution samples.
[0007] Furthermore, the second AI is trained through a variety of tatami design plan samples and corresponding scores, and the scores are determined by a third AI, and the third AI determines the scores based on the comment content, comment tone and comment expression corresponding to the various tatami design plan samples.
[0008] Furthermore, the third AI obtains the comment content, determines the positive information and negative information in the comment content, and calculates a first score. The third AI also obtains the comment voice, recognizes the comment tone, and calculates a second score. The third AI also obtains the commentator image, recognizes the comment expression, and calculates a third score. The score is the product of the first score, the second score, and the third score.
[0009] According to another aspect of the present invention, there is also provided a device for automatically designing tatami according to a housing type, including: an acquisition module for acquiring housing type parameters and user requirements of a target housing type; a generation module for generating multiple tatami design schemes according to the housing type parameters and the user requirements using a first AI; a scoring module for scoring the multiple tatami design schemes using a second AI; and a selection module for selecting the tatami design scheme having a score higher than a predetermined threshold.
[0010] Furthermore, the apartment parameters include apartment size and door and window positions, and the user requirements include design style and design color.
[0011] Furthermore, the first AI is trained through samples of multiple apartment parameters and user requirements and corresponding tatami design solution samples.
[0012] Furthermore, the second AI is trained through a variety of tatami design plan samples and corresponding scores, and the scores are determined by a third AI, and the third AI determines the scores based on the comment content, comment tone and comment expression corresponding to the various tatami design plan samples.
[0013] Furthermore, the third AI obtains the comment content, determines the positive information and negative information in the comment content, and calculates a first score. The third AI also obtains the comment voice, recognizes the comment tone, and calculates a second score. The third AI also obtains the commentator image, recognizes the comment expression, and calculates a third score. The score is the product of the first score, the second score, and the third score.
[0014] The present invention has at least the following beneficial effects: The present invention generates multiple tatami design plans using a first AI according to the housing parameters of the target housing type and user needs, uses a second AI to score them, and then selects the tatami design plan based on the scores; the present invention can overcome the defects of the prior art tatami design that is time-consuming, labor-intensive, and error-prone, and improves the user experience.
[0015] Other advantages, objectives and features of the present invention will be embodied in part through the following description, and in part will be understood by those skilled in the art through study and practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS
[0016] Figure 1-3 Rules that must be met by the tatami design sample in one embodiment of this application; Figure 4-9 The tatami design method generated for this application, with an original owner's floor plan on the left, a tatami layout automatically calculated by AI based on the cabinets in the middle, and the final rendering calculated by AI on the right. DETAILED DESCRIPTION
[0017] The present invention is further described in detail below in conjunction with the accompanying drawings so that those skilled in the art can implement the invention with reference to the description.
[0018] It should be understood that the terms such as "having", "including" and "comprising" used in the embodiments of the present application do not exclude the existence or addition of one or more other elements or their combinations. All directional indications (such as up, down, left, right, front, back...) in the embodiments of the present application are only used to explain the relative position relationship, movement, etc. between the components in a certain specific posture. If the specific posture changes, the directional indication will also change accordingly. When an element is referred to as "fixed on" or "set on" another element, it can be directly on the other element or there may be a centering element at the same time. When an element is referred to as "connecting" another element, it can be directly connected to another element or it can be indirectly connected to another element through a centering element. The description of "first", "second", etc. in the embodiments of the present application is only for descriptive purposes, and cannot be understood as indicating or implying its relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" can explicitly or implicitly include at least one of the features.
[0019] It should be noted that the technical solutions between the various embodiments of the present application can be combined with each other, but it must be based on the fact that ordinary technicians in the field can implement it. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such combination of technical solutions does not exist and is not within the scope of protection required by this application.
[0020] The embodiment of the present application provides a method for automatically designing tatami according to the house type, including: S1: Obtain the target apartment parameters and user requirements; For example, the apartment parameters include apartment size, door and window locations, corridor locations, etc., and the user requirements include design style, design color scheme, functional requirements, etc. The design style can be new Chinese style, Nordic minimalist style, minimalist luxury style, etc.; S2: Generate multiple tatami design plans according to the apartment parameters and the user needs using the first AI; For example, the tatami design plan includes the tatami layout style, ceiling, wall, floor hard decoration, cushions, accessories and other soft decorations; Exemplarily, a first AI is constructed using a generative adversarial network to generate a tatami design solution; S3: Using a second AI to score the plurality of tatami design solutions; Exemplarily, the second AI inputs one or more views in the tatami design plan and obtains a score; S4: Selecting the tatami design scheme having a score higher than a predetermined threshold; Exemplarily, multiple tatami design schemes are arranged from high to low according to the scores, and the top 20% of the tatami design schemes are selected for the user to choose, that is, the threshold is determined according to the actual score; This embodiment uses the first AI to generate multiple tatami design plans based on the apartment parameters of the target apartment and user needs, uses the second AI to score them, and then selects the tatami design plan based on the scores, thereby overcoming the defects of the prior art tatami design that is time-consuming, labor-intensive, and error-prone, and improving the user experience.
[0021] In another embodiment, the first AI is trained by using a plurality of samples of apartment parameters and user requirements and corresponding tatami design solution samples. The following lists the tatami design solution samples used in this embodiment, see Figure 1-3 ; (1) Style A: On the left side of the tatami is a desk with a wall cabinet, and below the tatami is a wardrobe with a hollow bottom (legs can be extended), which rests on the tatami; (2) Style B: On the left side of the tatami is a desk with a wall cabinet, and below the tatami is a wardrobe with a solid bottom, which rests on the tatami.
[0022] (3) Style C: On the left side of the tatami is a desk without a wall cabinet, and below the tatami is a wardrobe with a hollow bottom (legs can be extended), which rests on the tatami.
[0023] (4) Style D: On the left side of the tatami is a desk without a wall cabinet, and below the tatami is a wardrobe with a non-hollow bottom, which falls on the tatami.
[0024] (5) Style E tatami is a corner desk wardrobe without wall cabinets on the left. Below the tatami is a wardrobe with a hollow bottom (legs can be extended), and the wardrobe falls on the tatami.
[0025] (6) Style F tatami is a corner desk wardrobe without wall cabinets on the left. Below the tatami is a wardrobe with a non-hollow bottom, and the wardrobe falls on the tatami.
[0026] (7) Style G tatami is a corner desk wardrobe with a wall cabinet on the left, and below the tatami is a wardrobe with a hollow bottom (legs can be extended). The wardrobe falls on the tatami, and below the tatami is a floor-standing wardrobe on the left; (8) Style H tatami is a corner desk wardrobe with a wall cabinet on the left, and a wardrobe with a non-hollow bottom under the tatami. The wardrobe falls on the tatami, and there is a floor-standing wardrobe at the lower left of the tatami. (9) Style I: The tatami is a desk with a hanging cabinet above it, and there is a wardrobe under the tatami, with the wardrobe falling on top of the tatami; (10) Style J tatami has a desk without a wall cabinet on top and a wardrobe under the tatami, with the wardrobe resting on top of the tatami. (11) Style K tatami is a desk with a hanging cabinet on top, and a wardrobe on the left side of the tatami, which is located on top of the tatami; (12) Style L tatami is a desk without a cabinet above it, with a wardrobe on the left side of the tatami, which falls on top of the tatami; The above tatami design sample must also meet the following rules: 1. Rule 1: All cabinets added to the room cannot block doors or windows.
[0027] 2. Rule 2 Tatami depth range: The optimal range is 1200mm-1800mm, the applicable range is >900mm, and the tatami width range is >2000mm.
[0028] 3. Rule 3 Desk width range: The optimal range is 1000mm-1500mm, and the applicable range is >800mm.
[0029] 4. Rule 4: When the size of the desk is sufficient, it is preferred to use a desk with a wall cabinet, but the wall cabinet of the desk cannot block the window.
[0030] 5. Rule 5 The minimum length of the sleeping part of the tatami is 2000mm, the optimal width is 1200mm-1800mm, and the minimum width is 900mm.
[0031] 6. Rule 6: Width range of floor-standing wardrobes: The optimal range is >450mm, the applicable range is >300mm, the depth range is 500mm-600mm, and the back of the wardrobe should be against the wall.
[0032] 7. Rule 7 After ensuring the aisle size, as many cabinets as possible should be placed in the room to make full use of the storage space.
[0033] 8. Rule 8. After determining the tatami layout style, stretch the cabinets. The optimal depth of the wardrobe is 550mm, and the stretching range is 500mm-600mm. If the desk and tatami are both within the optimal range, stretch the width of the desk and the width (or depth) of the tatami in equal proportions. If one of the desk and the tatami is within the optimal range and the other is not, stretch the cabinet within the optimal range. If neither the desk nor the tatami is within the optimal range, stretch them in equal proportions while ensuring that both are within the applicable range.
[0034] 9. First, find out the walls where the doors and windows are located, Wall A and Wall C. According to Rules 1 and 2, it can be determined that the walls where tatami can be placed are Wall C and Wall D.
[0035] 10. Try wall C and wall D separately, place tatami on wall C, and give priority to finding the one with the optimal size. The styles that meet the requirements are A, B, C, and D. According to rule 4, styles A and B are better. According to rule 5, the best solution is style A. Once you find the style within the optimal range, there is no need to try wall D.
[0036] In another embodiment, the second AI is trained by a plurality of tatami design solution samples and corresponding scores, and the scores are determined by a third AI, which determines the scores according to the review content, review tone and review expression corresponding to the plurality of tatami design solution samples; Exemplarily, the second AI can be constructed based on a convolutional neural network, input multiple images of tatami design solutions, and obtain scores; In this step, the score used for the second AI training is determined by the third AI, and the score is determined based on the comment content, comment tone and comment expression, that is, the user's comment process on various tatami design plan samples is collected to obtain the comment content, comment audio, comment image, and then the comment content, comment tone and comment expression are obtained for the training of the second AI.
[0037] In another embodiment, the third AI obtains the comment content, determines the positive information and the negative information in the comment content, and calculates a first score. The third AI also obtains the comment voice, identifies the comment tone, and calculates a second score. The third AI also obtains the commentator image, identifies the comment expression, and calculates a third score. The score is the product of the first score, the second score, and the third score. Exemplarily, the third AI includes a language model, a sound model, and an image recognition model, for implementing the determination of the first score, the second score, and the third model; Exemplarily, the positive information may be a positive keyword, and the negative information may be a negative keyword. The third AI performs word segmentation on the comment content to determine the keyword. According to a preset scoring rule, different keywords have different scores. The scores of various keywords are added together to obtain a first score, and the value of the first score is between 0 and 100. Exemplarily, the comment tone includes a strong tone, a neutral tone and a negative tone, and the second score is between 0 and 1, and the score gradually decreases from strong to negative; the tone is determined by extracting the audio features of the comment speech and inputting them into a preset model; Exemplarily, the comment expressions include appreciation, general and derogatory, and the third score is between 0 and 1, and the score gradually decreases from appreciation to derogatory; the expression distinction is determined by extracting facial features and inputting them into a preset model; For example, after providing a tatami design plan to a user, the user is asked to comment, and the comment content, comment audio, and comment image are collected; Exemplarily, the third AI continuously extracts comment content, comment audio, and comment image, obtains a first score, a second score, a third score, and a score value, and trains and updates the second AI; In this step, the score is the product of the first score, the second score and the third score, which takes into account the reviewer's comment content, comment tone and comment expression, and can reflect the reviewer's true attitude to obtain a true score.
[0038] The embodiment of the present application also provides a device for automatically designing tatami according to the apartment type, comprising: an acquisition module, used to acquire the apartment type parameters and user requirements of the target apartment type; a generation module, used to generate multiple tatami design schemes according to the apartment type parameters and the user requirements using a first AI; a scoring module, used to score the multiple tatami design schemes using a second AI; a selection module, used to select the tatami design scheme with a score higher than a predetermined threshold; The acquisition module, generation module, scoring module and selection module of this embodiment are composed of programs. The acquisition module generates multiple tatami design plans based on the apartment parameters and user needs of the target apartment type, the generation module uses the first AI to generate multiple tatami design plans, the scoring module uses the second AI to score, and then the selection module selects the tatami design plan based on the score, which overcomes the defects of the prior art tatami design that is time-consuming, labor-intensive and error-prone, and improves the user experience.
[0039] In another embodiment, the first AI is trained through samples of multiple apartment parameters and user requirements and corresponding tatami design solution samples; illustratively, the generation of the tatami design solution is achieved using a generative adversarial network (GAN).
[0040] In another embodiment, the second AI is trained by a plurality of tatami design solution samples and corresponding scores, and the scores are determined by a third AI, which determines the scores according to the review content, review tone and review expression corresponding to the plurality of tatami design solution samples; Exemplarily, the second AI may use a convolutional neural network, input multiple images of tatami design solutions, and obtain scores; In this step, the score used for the second AI training is determined by the third AI, and the score is determined based on the comment content, comment tone and comment expression, that is, the user's comment process on various tatami design plan samples is collected to obtain the comment content, comment audio, comment image, and then the comment content, comment tone and comment expression are obtained for the training of the second AI.
[0041] In another embodiment, the third AI obtains the comment content, determines the positive information and the negative information in the comment content, and calculates a first score. The third AI also obtains the comment voice, identifies the comment tone, and calculates a second score. The third AI also obtains the commentator image, identifies the comment expression, and calculates a third score. The score is the product of the first score, the second score, and the third score. Exemplarily, the positive information may be positive keywords, and the reverse information may be reverse keywords. The third AI performs word segmentation on the comment content to determine the keywords. According to the preset scoring rules, different keywords have different scores. The scores of various keywords are added together to obtain a first score. The comment tone includes strong tone, neutral tone and negative tone. The second score is between 0-1, and the score gradually decreases from strong to negative. The tone is distinguished by extracting the audio features of the comment voice and inputting them into a preset model. The comment expressions include appreciation, general and derogatory. The third score is between 0-1, and the score gradually decreases from appreciation to derogatory. The expression distinction is determined by extracting facial features and inputting them into a preset model. The score is the product of the first score, the second score and the third score, taking into account the comment content, comment tone and comment expression of the commentator, and can reflect the commentator's true attitude to obtain a true score. After testing, compared with considering only the comment content, users' satisfaction with the ranking of tatami design scheme samples increased by 20%.
[0042] The following Figure 5-9 The generated tatami design plan is used as an example for explanation.
[0043] like Figure 5 Shown, tatami style D.
[0044] 1. First, find the walls where the doors and windows are located, which are walls B and D in the picture. According to rule 1, determine that the walls where tatami can be placed are C and D.
[0045] 2. Try it on wall C, giving priority to the optimal size. According to the optimal sizes of tatami, desk, and floor-standing wardrobe, the tatami styles that can be placed are styles A, B, C, and D.
[0046] 3. According to rule 4, the desk ceiling cannot block the window, so remove styles A and B.
[0047] 4. According to Rule 5, the length of the tatami for sleeping people must be greater than 2000mm, which is met by both Style C and Style D.
[0048] 5. According to Rule 9, under the same conditions, the style with more cabinets and better storage space is given priority, so choose Style D. Since Style D is already within the optimal size range, stop trying and finally choose Style D.
[0049] 6. First, AI automatically calculates the corresponding ceiling, wall, floor and other hard furnishings, as well as tatami mats and accessories based on the selected Nordic style and the calculated tatami layout style, to ensure the overall beauty and coordination. Second, AI-rated designs can be found in Figure 5 .
[0050] like Figure 6 Shown: Tatami Style E.
[0051] 1. First, find the walls where the doors and windows are located, which are walls B and D in the picture. According to rule 1, determine that the walls where tatami can be placed are C and D.
[0052] 2. Try it on wall C, giving priority to the optimal size. According to the optimal sizes of tatami, desk, and floor-standing wardrobe, the tatami styles that can be placed are styles A, B, C, D, E, and F.
[0053] 3. According to rule 7, give priority to solutions with more cabinets and better storage space utilization. So remove A, B, C, and D, leaving only solutions E and F.
[0054] 7. According to Rule 5, the length of the tatami for a person to sleep on must be greater than 2000mm. The wall length here is 2450mm. After deducting the minimum depth of the wardrobe of 500mm, the size is less than 2000mm, so plan F is eliminated and plan E is finally selected. Since E is already within the optimal size range, stop trying.
[0055] 8. First AI automatically calculates the corresponding ceiling, wall, floor and other hard decorations, as well as tatami mats and accessories, based on the selected girl's room style and the calculated tatami layout style, to ensure the overall beauty and coordination. Second AI's higher-scoring designs can be found in Figure 6 .
[0056] like Figure 7Shown: Tatami Style H.
[0057] 1. First, find the walls where the doors and windows are located, which are walls A and C in the picture. According to rule 1, determine that the walls where tatami can be placed are C and D.
[0058] 2. Try it on wall B, giving priority to the optimal size. According to the optimal sizes of tatami, desk, and floor-standing wardrobe, the tatami styles that can be placed are styles A, B, C, D, E, F, G, and H.
[0059] 3. According to rule 7, give priority to solutions with more cabinets and better storage space utilization. So after removing A, B, C, D, E, and F, only solutions G and H are left.
[0060] 4. According to Rule 5, the length of the tatami for a person to sleep on must be greater than 2000mm. The wall length here is 2600mm. After deducting the optimal depth of the wardrobe of 550mm, the size of 2100mm meets the requirements, so plan G is eliminated and plan H is finally selected. Since H is already within the optimal size range, stop trying.
[0061] 5. First, AI automatically calculates the corresponding ceiling, wall, floor and other hard furnishings, as well as tatami mats and accessories based on the selected French style and the calculated tatami layout style, to ensure the overall beauty and coordination. Second, AI-rated designs can be found in Figure 7 .
[0062] like Figure 8 Shown: Tatami Style J.
[0063] 1. First, find the walls where the doors and windows are located, walls A and C in the picture. According to rule 1, determine that the walls where tatami can be placed are B and C.
[0064] 2. Try it on wall B, giving priority to the optimal size. According to the optimal sizes of tatami, desk, and floor-standing wardrobe, the tatami styles that can be placed are styles I, J, K, and L.
[0065] 3. According to rule 4, the desk and wall cabinet cannot block the window, so remove style I and style K, leaving style J and style L.
[0066] 4. Try to put down style J and style L, and stretch them according to rule 8. The depth of style J wardrobe is 550mm (optimal). The length of the tatami sleeping part is 2000mm, so the size of the tatami is 2550mm*1600mm, and the size of the tatami sleeping part is 2000mm*1600mm. The tatami and desk need to be stretched all over the wall, so the size of the desk is 3440mm-2550mm=890mm. The depth of style L wardrobe is 550mm, the minimum tatami is 2000mm, and the minimum width of the desk is 1000mm. After the wall length is 3440mm, the desk width is 1147mm, the tatami width is 2293mm, the tatami size is 2293mm*1600mm, and the size of the tatami sleeping part is 2293mm*1050mm. According to the weight comparison, the tatami sleeping person in style J has the optimal value, and the desk deviates from the optimal value by 0.01. The desk in style L is at the optimal value, and the width of the tatami sleeping person deviates from the optimal value by 0.125. So style J is selected.
[0067] 6. Style J of wall B has a value that is not optimal, so continue to try wall C. All styles have desks or wardrobes blocking the interior doors and cannot be used.
[0068] 7. Finally, determine the B wall and apply style J.
[0069] 8. First AI automatically calculates the corresponding ceiling, wall, floor and other hard furnishings, as well as tatami mats and accessories, based on the selected boy's room style and the calculated tatami layout style, to ensure the overall beauty and coordination. Second AI's designs with higher scores can be found in Figure 8 .
[0070] like Fig. 9 Shown: Tatami Style K.
[0071] 1. First, find the walls where the doors and windows are located, which are walls B and D in the picture. According to rule 1, determine that the walls where tatami can be placed are B and C.
[0072] 2. Try it on wall B. The minimum width of tatami is 2000mm and the length of wall B is 1900mm, so the tatami style cannot be placed.
[0073] 3. Try it on wall C. Styles A, B, C, D, E, F, G, and H all have cabinets blocking the interior doors and cannot be used. Styles I and J have cabinets blocking the windows and cannot be used. The only styles left are K and L.
[0074] 4. According to rule 4, it is preferred to use a desk with a wall cabinet, so choose K. Finally, it is decided to use style K on the C wall.
[0075] 5. First, AI automatically calculates the corresponding ceiling, wall, floor and other hard furnishings, as well as tatami mats and accessories, based on the selected new Chinese style and the calculated tatami layout style, to ensure the overall beauty and coordination. Second, AI-rated designs refer to Fig. 9 .
[0076] Although the embodiments of the present invention have been disclosed as above, they are not limited to the applications listed in the specification and the implementation modes, and they can be fully applied to various fields suitable for the present invention. For those familiar with the art, additional modifications can be easily implemented. Therefore, without departing from the general concept defined by the claims and the scope of equivalents, the present invention is not limited to the specific details and the illustrations shown and described herein.
Claims
1. A method for automatically designing tatami according to a house type, characterized in that: include: S1: Obtain the target apartment parameters and user requirements; S2: Generate multiple tatami design plans according to the apartment parameters and the user needs using the first AI; S3: Using a second AI to score the plurality of tatami design solutions; S4: Select the tatami design scheme with a score higher than a predetermined threshold.
2. The method for automatically designing tatami according to the house type according to claim 1, characterized in that: The apartment parameters include apartment size and door and window positions, and the user requirements include design style and design color.
3. The method for automatically designing tatami according to house type as claimed in claim 1, characterized in that: The first AI is trained through samples of multiple apartment parameters and user requirements and corresponding tatami design plan samples.
4. The method for automatically designing tatami according to house type as claimed in claim 1, characterized in that: The second AI is trained through a variety of tatami design plan samples and corresponding scores, and the scores are determined by a third AI, which determines the scores based on the comment content, comment tone and comment expression corresponding to the various tatami design plan samples.
5. The method for automatically designing tatami according to the house type as claimed in claim 4, characterized in that: The third AI obtains the comment content, determines the positive information and negative information in the comment content, and calculates a first score. The third AI also obtains the comment voice, recognizes the comment tone, and calculates a second score. The third AI also obtains the commentator image, recognizes the comment expression, and calculates a third score. The score is the product of the first score, the second score, and the third score.
6. A device for automatically designing tatami according to the house type, characterized in that: include: An acquisition module is used to obtain the housing parameters and user requirements of the target housing type; A generation module, configured to generate a plurality of tatami design solutions according to the apartment parameters and the user requirements using the first AI; A scoring module, used for scoring the plurality of tatami design schemes using a second AI; The selection module is used to select the tatami design scheme with a score higher than a predetermined threshold.
7. The device for automatically designing tatami according to the house type as claimed in claim 6, characterized in that: The apartment parameters include apartment size and door and window positions, and the user requirements include design style and design color.
8. The device for automatically designing tatami according to the house type as claimed in claim 6, characterized in that: The first AI is trained through samples of multiple apartment parameters and user requirements and corresponding tatami design plan samples.
9. The device for automatically designing tatami according to house type as claimed in claim 6, characterized in that: The second AI is trained through a variety of tatami design plan samples and corresponding scores, and the scores are determined by a third AI, which determines the scores based on the comment content, comment tone and comment expression corresponding to the various tatami design plan samples.
10. The device for automatically designing tatami according to house type as claimed in claim 9, characterized in that: The third AI obtains the comment content, determines the positive information and negative information in the comment content, and calculates a first score. The third AI also obtains the comment voice, recognizes the comment tone, and calculates a second score. The third AI also obtains the commentator image, recognizes the comment expression, and calculates a third score. The score is the product of the first score, the second score, and the third score.
Citation Information
Patent Citations
Data processing method, apparatus and computer device based on satisfaction evaluation
CN109460891A
Agricultural product user score prediction method
CN111652416A
Urban scenic spot intelligent recommendation system and method thereof
CN112148985A
Method and device for analyzing authenticity of video comments, electronic equipment and storage medium
CN113065052A
Interior design scheme acquisition method and device, computer equipment and storage medium
CN113505429A
