Data acquisition and analysis system and method for interior decoration

Abstract

The invention relates to the technical field of data acquisition, and more specifically discloses a data acquisition and analysis system and method for indoor decoration, and the method comprises the steps: extracting the decoration scheme information of a target indoor decoration project and pre-estimating the characteristics of parameter information, constructing a related feature map, and decoding the related feature map, according to the method, the material consumption of the indoor decoration engineering is accurately estimated, so that the problems of low accuracy, material waste or construction period delay existing in a single means based on manual estimation at present are effectively solved.

Description

Technical Field

[0001] The present application relates to the field of data acquisition technology, and more specifically, to a data acquisition and analysis system and method for interior decoration. Background Art

[0002] Interior decoration refers to the design and renovation of the interior space of a house to make it more suitable for living and use. Interior decoration involves a wide range of content, including walls, floors, ceilings, doors and windows, lighting, furniture, soft furnishings and other aspects.

[0003] Interior decoration is divided into hard furnishings and soft furnishings. Hard furnishings require a vast array of materials, and the types of materials involved are also numerous. Therefore, estimating interior decoration material consumption is a complex and crucial process. Failure to accurately estimate interior decoration material consumption not only delays the project schedule, but also results in material waste and, in severe cases, compromises the quality of the interior decoration construction. Therefore, accurate estimation of interior decoration material consumption is crucial.

[0004] However, the existing methods for estimating the consumption of engineering materials for interior decoration are mostly based on manual inspection and rough calculation of the project volume of the target decoration project, so as to estimate the consumption of engineering materials for interior decoration. The means of evaluating the consumption of engineering materials are relatively simple and have low accuracy. In addition, overestimation leads to excess engineering materials, resulting in material waste, while underestimation leads to insufficient engineering materials, which delays the progress of the project.

[0005] Therefore, a data collection and analysis system and method for interior decoration are desired. Summary of the Invention

[0006] To address the above technical issues, the present application provides a data collection and analysis system and method for interior decoration. By extracting features of interior decoration schemes and influencing parameters, constructing and decoding associated feature maps, and accurately estimating the material consumption of interior decoration projects, the present application provides a data collection and analysis system and method for interior decoration.

[0007] Accordingly, according to one aspect of the present application, a data collection and analysis system for interior decoration is provided, comprising:

[0008] An interior decoration information collection module, configured to obtain decoration scheme information and estimated impact parameter information of a target interior decoration project, wherein the decoration scheme information of the target interior decoration project includes information on the categories of engineering materials required for the corresponding decoration scheme, and the estimated impact parameter information of the target interior decoration project includes the house area and spatial layout;

[0009] an interior decoration information processing module, configured to extract a decoration scheme information semantic feature vector and an estimated impact parameter information association feature vector from the decoration scheme information of the target interior decoration project and the estimated impact parameter information of the target interior decoration project, respectively;

[0010] an interior decoration information fusion module, configured to construct an estimated engineering material consumption correlation feature vector between the decoration scheme information semantic feature vector and the estimated influencing parameter information correlation feature vector, and perform kernel co-modulation based on intrinsic manifold regularization on the estimated engineering material consumption correlation feature vector to obtain a modulated estimated engineering material consumption correlation feature vector;

[0011] The interior decoration information analysis module is used to pass the modulated engineering material consumption estimate associated feature vector through a decoder to obtain a decoded value, and the decoded value is used to represent the final engineering material consumption estimate of the target interior decoration project.

[0012] According to another aspect of the present application, a data collection and analysis method for interior decoration is provided, comprising:

[0013] Obtaining decoration scheme information and estimated impact parameter information of a target interior decoration project, wherein the decoration scheme information of the target interior decoration project includes information on the categories of engineering materials required for the corresponding decoration scheme, and the estimated impact parameter information of the target interior decoration project includes house area and space layout;

[0014] extracting a decoration scheme information semantic feature vector and an estimated impact parameter information association feature vector from the decoration scheme information of the target interior decoration project and the estimated impact parameter information of the target interior decoration project respectively;

[0015] Constructing an estimated engineering material consumption correlation feature vector between the decoration scheme information semantic feature vector and the estimated influencing parameter information correlation feature vector, and performing kernel co-modulation based on intrinsic manifold regularization on the estimated engineering material consumption correlation feature vector to obtain a modulated estimated engineering material consumption correlation feature vector;

[0016] The modulated engineering material consumption estimate associated feature vector is passed through a decoder to obtain a decoded value, and the decoded value is used to represent the final engineering material consumption estimate of the target interior decoration project.

[0017] Compared with the existing technology, the present application provides a data collection and analysis system and method for interior decoration, which extracts the characteristics of the decoration scheme information and estimated influencing parameter information of the target interior decoration project, constructs and decodes the associated feature map to achieve accurate estimation of the material consumption of the interior decoration project, thereby effectively solving the problems of low accuracy, waste of materials or delay in construction period existing in the current single means based on manual estimation. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The above and other purposes, features, and advantages of the present application will become more apparent through a more detailed description of the embodiments of the present application in conjunction with the accompanying drawings. The accompanying drawings are intended to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the drawings, the same reference numerals generally represent the same components or steps.

[0019] Figure 1 Schematic diagram of a data acquisition and analysis system for interior decoration according to an embodiment of the present application.

[0020] Figure 2 Schematic diagram of a block diagram of an interior decoration information processing module in a data acquisition and analysis system for interior decoration according to an embodiment of the present application.

[0021] Figure 3 Schematic diagram of a decoration scheme information extraction unit in a data acquisition and analysis system for interior decoration according to an embodiment of the present application.

[0022] Figure 4 Schematic diagram of a block diagram of an estimated influencing parameter information extraction unit in a data acquisition and analysis system for interior decoration according to an embodiment of the present application.

[0023] Figure 5 Flowchart of a data collection and analysis method for interior decoration according to an embodiment of the present application. DETAILED DESCRIPTION

[0024] Various exemplary embodiments, features, and aspects of the present application will be described in detail below with reference to the accompanying drawings. The same reference numerals in the accompanying drawings represent elements with the same or similar functions. Although various aspects of the embodiments are shown in the accompanying drawings, the drawings are not necessarily drawn to scale unless otherwise indicated.

[0025] The word “exemplary” is used exclusively herein to mean “serving as an example, example, or illustration.” Any embodiment described herein as “exemplary” is not necessarily to be construed as preferred or advantageous over other embodiments.

[0026] In addition, numerous specific details are provided in the following detailed description to better illustrate the present application. Those skilled in the art will appreciate that the present application can be practiced without certain specific details. In some instances, methods, means, components, and circuits well known to those skilled in the art are not described in detail in order to highlight the main purpose of the present application.

[0027] Furthermore, 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 defined as "first" or "second" may explicitly or implicitly include one or more of the features. Throughout the description of this application, "plurality" means two or more, unless otherwise specifically defined.

[0028] Figure 1 The figure shows a block diagram of a data collection and analysis system for interior decoration according to an embodiment of the present application. Figure 1 As shown, according to an embodiment of the present application, the data acquisition and analysis system 100 for interior decoration includes: an interior decoration information acquisition module 110, which is used to obtain decoration scheme information and estimated impact parameter information of a target interior decoration project, wherein the decoration scheme information of the target interior decoration project includes the category information of engineering materials required for the corresponding decoration scheme, and the estimated impact parameter information of the target interior decoration project includes the house area and space layout; an interior decoration information processing module 120, which is used to extract the decoration scheme information semantic feature from the decoration scheme information of the target interior decoration project and the estimated impact parameter information of the target interior decoration project respectively. The interior decoration information fusion module 130 is used to construct an engineering material consumption estimation association feature vector between the decoration scheme information semantic feature vector and the estimated influencing parameter information association feature vector, and perform kernel co-modulation based on intrinsic manifold regularization on the engineering material consumption estimation association feature vector to obtain a modulated engineering material consumption estimation association feature vector; the interior decoration information analysis module 140 is used to pass the modulated engineering material consumption estimation association feature vector through a decoder to obtain a decoded value, which is used to represent the final estimated engineering material consumption of the target interior decoration project.

[0029] In this embodiment of the present application, the interior decoration information acquisition module 110 is configured to acquire decoration scheme information and estimated influencing parameter information for a target interior decoration project. The decoration scheme information includes information on the categories of engineering materials required for the corresponding decoration scheme, and the estimated influencing parameter information includes information on the floor area and spatial layout. It should be understood that the decoration scheme information provides information on the categories of engineering materials required for the corresponding decoration scheme, including specific material types, quantities, and specifications. This clear information helps accurately determine the consumption of each material, thereby achieving a precise estimate. The estimated influencing parameter information includes parameters such as floor area and spatial layout, which directly impact the material consumption of the decoration project. Floor area determines the total amount of materials required for the overall decoration, while spatial layout affects the specific material consumption in different areas or locations. By combining decoration scheme information and influencing parameter information, an estimate can be made based on actual data, avoiding errors caused by subjective estimation, thereby improving the accuracy and reliability of the estimate. Systematized analysis using technologies such as artificial intelligence can comprehensively process and analyze large amounts of data, resulting in more accurate estimates of project material consumption and reducing uncertainty. Based on the information obtained, the estimation model can be adjusted and optimized at any time to adapt to new situations or changes that may arise, ensuring the timeliness and accuracy of the estimation results.

[0030] In an embodiment of the present application, the interior decoration information processing module 120 is used to extract the decoration scheme information semantic feature vector and the estimated impact parameter information association feature vector from the decoration scheme information of the target interior decoration project and the estimated impact parameter information of the target interior decoration project, respectively. It should be understood that converting the decoration scheme information and the estimated impact parameter information into feature vectors can represent the information in a unified numerical form, which is convenient for machine learning algorithm processing and analysis. Feature vector extraction can extract and abstract complex features in the original information, thereby reducing the information dimension, reducing redundant information, and improving the efficiency of data processing. Through feature vector extraction, the semantic features and association information in the decoration scheme information and the estimated impact parameter information can be captured, which helps to better understand and analyze the data. The extracted feature vector can be used as the input of the machine learning model to establish a prediction model or perform data analysis, thereby achieving an accurate estimate of the consumption of engineering materials. After the information is converted into a feature vector, data visualization and exploratory data analysis can be performed to help users understand the characteristics and correlation of the data more intuitively.

[0031] Specifically, in one embodiment of the present application, Figure 2 FIG2 is a block diagram of an interior decoration information processing module in a data collection and analysis system for interior decoration according to an embodiment of the present application. Figure 2As shown, in the above-mentioned data acquisition and analysis system 100 for interior decoration, the interior decoration information processing module 120 includes: a decoration scheme information extraction unit 121, which is used to perform convolution encoding on the decoration scheme information of the target interior decoration project to obtain a semantic feature vector of the decoration scheme information; and an estimated impact parameter information extraction unit 122, which is used to perform word segmentation on the estimated impact parameter information of the target interior decoration project and then encode it to obtain an associated feature vector of the estimated impact parameter information.

[0032] Specifically, the decoration scheme information extraction unit 121 is configured to perform convolution coding on the decoration scheme information of the target interior decoration project to obtain a semantic feature vector for the decoration scheme information. It should be understood that convolution coding is an effective feature extraction method that can extract important features from decoration scheme information, capture key information and patterns between different decoration schemes, and help improve feature representation capabilities. The convolution operation can effectively capture local correlations in data and has excellent extraction capabilities for local structures and patterns that may exist in the decoration scheme information, helping to retain important semantic information. Convolutional neural networks have the characteristic of parameter sharing, which can reduce the number of model parameters, reduce the risk of overfitting, and improve the model's generalization ability, making them suitable for semantic feature extraction of decoration scheme information. The convolution operation is translation-invariant, which is effective for identifying features at different locations in the decoration scheme information and is not affected by position changes, facilitating the extraction of the overall semantic features of the decoration scheme information. Convolutional neural networks have been widely used and successfully in fields such as image processing. They have good versatility and adaptability and can be effectively applied to the processing and feature extraction of decoration scheme information.

[0033] further, Figure 3 The figure shows a schematic diagram of a decoration scheme information extraction unit in a data collection and analysis system for interior decoration according to an embodiment of the present application. Figure 3 As shown, in the interior decoration information processing module 120 of the above-mentioned data acquisition and analysis system 100 for interior decoration, the decoration scheme information extraction unit 121 includes: a semantic understanding subunit 1211, which is used to obtain a decoration scheme information feature vector from the decoration scheme information of the target interior decoration project using a semantic understanding model; and a first convolutional coding subunit 1212, which is used to pass the decoration scheme information feature vector through an information extractor based on a first convolutional neural network model to obtain a decoration scheme information semantic feature vector.

[0034] Specifically, the semantic understanding subunit 1211 is used to obtain a decoration scheme information feature vector from the decoration scheme information of the target interior decoration project using a semantic understanding model. It should be understood that the semantic understanding model can deeply understand the semantic information of the text data, help the system better understand the meaning and context of the decoration scheme information, and thus extract more representative and meaningful features. The semantic understanding model can effectively extract key features from the decoration scheme information, capture important information and semantic content in the text, and help improve the representation ability and discrimination of the features. The semantic understanding model can take into account the contextual information of the text data, model the complex relationships and context in the decoration scheme information, and extract feature vectors that are more coherent and consistent. The semantic understanding model can be adjusted and optimized according to specific tasks and data, has strong flexibility and adaptability, and can adapt to different types of decoration scheme information extraction needs. The semantic understanding model based on deep learning performs well in processing natural language text, can effectively capture semantic information in text data, and extract high-quality feature vectors.

[0035] Specifically, the semantic understanding sub-unit includes: a word vector conversion secondary sub-unit, which is used to use the word embedding layer of the semantic understanding model to map each word in the decoration scheme information of the target interior decoration project into a word vector to obtain a decoration scheme information word vector sequence; a semantic processing secondary sub-unit, which is used to use the Bert model of the semantic understanding model to process the decoration scheme information word vector sequence to obtain a decoration scheme information word feature vector sequence; and a context encoding secondary sub-unit, which is used to use the bidirectional LSTM network of the semantic understanding model to context encode the decoration scheme information word feature vector sequence to obtain the weather information feature vector.

[0036] Correspondingly, the word vector conversion secondary sub-unit is used to use the word embedding layer of the semantic understanding model to map each word in the decoration scheme information of the target interior decoration project into a word vector to obtain a decoration scheme information word vector sequence. It should be understood that word embedding technology can map words in the text into a low-dimensional vector space, so that each word can be represented by a dense vector. These vectors capture the semantic relationship between words and help to effectively represent the semantic information of the vocabulary. Word embedding can map words that appear in similar contexts to similar vector space positions, thereby capturing the contextual information between words, which helps to understand the semantic association and contextual meaning of words in the decoration scheme information. Word embedding maps high-dimensional sparse word representations into low-dimensional dense vector space, reducing the number of feature dimensions while retaining the semantic similarity between words, which is conducive to improving the efficiency and quality of feature representation. The word vector obtained by word embedding has a certain generalization ability, can capture the common semantic features in different decoration scheme information, and helps to extract representative and universal feature vectors. Combining word embedding with deep learning models can further improve the model's performance in processing decoration scheme information, enabling the system to better understand and process text data, providing strong support for subsequent feature extraction and information understanding.

[0037] Furthermore, the semantic processing secondary sub-unit is used to use the Bert model of the semantic understanding model to process the decoration scheme information word vector sequence to obtain the decoration scheme information word feature vector sequence. It should be understood that the Bert model is a pre-trained model based on the Transformer architecture, which can capture the contextual information in the text data and understand the relationship between words in the sentence, so as to better represent the semantics and context in the decoration scheme information. The Bert model can simultaneously consider the contextual information on the left and right sides of a word through bidirectional context modeling, effectively capture the relationship between the word and the surrounding words, and help to more comprehensively understand the meaning of the words in the decoration scheme information. The Bert model has been pre-trained on a large-scale corpus and has learned rich language representations. It can convert the word vector sequence in the decoration scheme information into a word feature vector sequence with more semantic information, thereby improving the representation ability of the features. Since the Bert model has been pre-trained on a large-scale corpus, it can be applied to the decoration scheme information processing task through fine-tuning or transfer learning, thereby improving the performance and generalization ability of the model on specific tasks. The Bert model has made remarkable achievements in the field of natural language processing. It can effectively capture complex semantic information in text data and convert it into high-quality feature representation, which helps to improve the feature extraction effect of decoration scheme information.

[0038] Furthermore, the context encoding secondary sub-unit is used to use the bidirectional LSTM network of the semantic understanding model to contextually encode the decoration scheme information word feature vector sequence to obtain the weather information feature vector. It should be understood that the bidirectional LSTM network can effectively capture the context information of the words in the decoration scheme information, and by learning the order and contextual relationship of the words in the sequence, it helps to more comprehensively understand the semantics and context of the words in the decoration scheme information. The LSTM network is a recurrent neural network suitable for sequence data. It can model sequence data and retain the order information of the words appearing in the sequence, which is conducive to capturing the sequence features and context information in the decoration scheme information. The LSTM network can effectively handle long-distance dependencies through the design of gate units, avoid the gradient disappearance or gradient explosion problems in traditional RNN networks, and help capture long-distance semantic associations in the decoration scheme information. By encoding the word feature vector sequence of the decoration scheme information through the bidirectional LSTM network, a richer and more abstract feature representation can be obtained, and a feature vector sequence with more semantic information can be extracted, which helps to characterize the complex semantic features in the decoration scheme information. The bidirectional LSTM network can simultaneously consider the contextual information on the left and right sides of the word, comprehensively considering the context of the entire sentence, and helps to integrate the features of each word into a more global feature representation, thereby obtaining a more accurate and comprehensive weather information feature vector.

[0039] Specifically, the first convolutional encoding subunit 1212 is configured to pass the decoration scheme information feature vector through an information extractor based on a first convolutional neural network model to obtain a decoration scheme information semantic feature vector. It should be understood that convolutional neural networks can effectively capture local features when processing text data. By sliding the convolution kernel across different locations to extract features, this helps identify local semantic features in decoration scheme information, such as phrases and word groups. Convolutional neural networks are translationally invariant, meaning that the same features appearing at different locations can be detected by the same convolution kernel. This helps extract semantic features in decoration scheme information that are not affected by position, thereby improving the generalization capability of features. Through the information extractor of the convolutional neural network model, the decoration scheme information feature vector can be subjected to multi-layer convolution and pooling operations to achieve feature fusion and abstraction, thereby obtaining a higher-level semantic feature representation and improving the representation capability of decoration scheme information. The parameter sharing mechanism in convolutional neural networks can reduce the number of model parameters, improve the model's computational efficiency and training speed, and also help improve the model's generalization capability in decoration scheme information processing tasks. Convolutional neural networks have achieved good results in fields such as image processing and natural language processing. Through the information extractor based on the first convolutional neural network model, the semantic features of decoration scheme information can be effectively extracted, thereby improving the performance and effect of the model in processing decoration scheme information.

[0040] Accordingly, in a specific example of the present application, the estimated impact parameter information extraction unit 122 is used to segment the estimated impact parameter information of the target interior decoration project and then encode it to obtain the estimated impact parameter information associated feature vector. It should be understood that segmenting the parameter information can better capture the semantic association between words, which helps to extract the specific meaning and characteristics of each parameter. After encoding, these segmented information can be converted into numerical feature vectors to better represent the semantics of the parameter information. Converting the segmented parameter information into a feature vector by encoding helps to extract key features in the parameter information, such as important parameters, associations between parameters, etc., and provide a more informative representation for subsequent analysis and processing. Converting the estimated impact parameter information into a feature vector by encoding can achieve unified representation and processing of different parameter information, which facilitates unified feature calculation and analysis in subsequent models. The encoded feature vector usually has a lower dimension, which is conducive to reducing the complexity and computational cost of the data, while retaining the key features in the parameter information and improving the efficiency and performance of the model. The encoded feature vector can be used as the input of the machine learning model to help the model better understand and process the estimated influencing parameter information, thereby accurately estimating and analyzing the impact of the target interior decoration project.

[0041] further, Figure 4 The figure shows a block diagram of the estimated impact parameter information extraction unit in the data acquisition and analysis system for interior decoration according to an embodiment of the present application. Figure 4 As shown, in the interior decoration information processing module 120 of the above-mentioned data acquisition and analysis system 100 for interior decoration, the estimated impact parameter information extraction unit 122 includes: a context encoding subunit 1221, which is used to perform word segmentation processing on the estimated impact parameter information of the target interior decoration project and then pass it through a first context encoder based on a converter to obtain multiple estimated impact parameter information feature vectors; a vector cascade subunit 1222, which is used to cascade the multiple estimated impact parameter information feature vectors to obtain an estimated impact parameter information association feature vector.

[0042] Specifically, the context encoding subunit 1221 is configured to perform word segmentation processing on the estimated impact parameter information of the target interior decoration project, and then pass it through a first context encoder based on a transformer to obtain multiple estimated impact parameter information feature vectors. It should be understood that word segmentation can better understand the meaning and semantics of each parameter information, allowing each parameter to be more accurately represented. The first context encoder based on a transformer can convert this segmented information into a semantically rich feature vector, better capturing the semantics and associations of the parameter information. The first context encoder helps capture the contextual associations of the parameter information, namely, the relationship between each parameter information and its surrounding parameters, facilitating a more comprehensive understanding of the connections and impacts between the parameter information. The feature vectors obtained by the transformer encoder contain more semantic information and contextual associations, providing a richer and more meaningful feature representation, thereby improving the representation capabilities of the estimated impact parameter information. Obtaining multiple estimated impact parameter information feature vectors provides a diverse feature representation, covering parameter information features at different levels and angles, facilitating a more comprehensive description of the impact parameters of the target interior decoration project. These multiple feature vectors can serve as input to the model, helping it better understand and process the parameter information of the interior decoration project, thereby achieving accurate impact estimation and analysis.

[0043] Specifically, the vector cascade subunit 1222 is used to cascade the multiple estimated impact parameter information feature vectors to obtain the estimated impact parameter information associated feature vector. It should be understood that cascading multiple feature vectors can fuse their feature dimensions together to form a more comprehensive and comprehensive feature vector. This helps to integrate information from different levels, angles or sources and improve the representation ability of the feature vector. Cascading multiple feature vectors can retain the information in each feature vector while merging them into a more comprehensive feature vector. This can make the resulting associated feature vector richer and more comprehensive, covering the information of different feature vectors. By cascading multiple feature vectors, the relationship and mutual influence between parameter information can be better captured. This helps to establish a connection model between parameter information and improve the understanding and representation of the correlation of parameter information. The resulting associated feature vector can be used as a higher-level feature representation, providing a more comprehensive and integrated data foundation for subsequent analysis and modeling. This helps to improve the system's comprehensive assessment capabilities of the impact of interior decoration projects. After multiple eigenvectors are cascaded to form an associated eigenvector, it can be used as the input of the model to help the model better understand and process the association between parameter information, thereby improving the accuracy and effect of the estimation of the impact on interior decoration projects.

[0044] In this embodiment of the present application, the interior decoration information fusion module 130 is configured to construct an estimated engineering material consumption correlation feature vector between the semantic feature vector of the decoration scheme information and the associated feature vector of the estimated influencing parameter information, and then perform kernel co-modulation based on intrinsic manifold regularization on the estimated engineering material consumption correlation feature vector to obtain the modulated estimated engineering material consumption correlation feature vector. It should be understood that by constructing the correlation feature vector, the relationship between the semantic feature vector of the decoration scheme information and the associated feature vector of the estimated influencing parameter information can be intuitively displayed. This facilitates a deeper understanding of the connection between the decoration scheme and parameter information, thereby better predicting engineering material consumption. The correlation feature vector provides an intuitive visualization, helping users intuitively understand the correlation between different feature vectors. This visualization facilitates rapid understanding and analysis of the relationship between decoration scheme information and parameter information. Through the correlation feature vector, correlations between the semantic feature vector of the decoration scheme information and the associated feature vector of the estimated influencing parameter information can be discovered. These correlations can provide important clues and references for estimating engineering material consumption, helping to improve the accuracy of consumption estimates. The correlation feature vector can be used for data exploration and analysis, helping to reveal potential relationships and patterns between different feature vectors. This helps uncover hidden information in the data and provides guidance for further data mining and analysis. By constructing associated eigenvectors, guidance can be provided for model optimization and improvement. Analyzing the relationships between eigenvectors can help optimize the model's feature selection and modeling process, thereby improving the accuracy and efficiency of engineering material consumption estimates.

[0045] Accordingly, in one embodiment of the present application, the interior decoration information fusion module 130 includes: a fusion matrix unit, used to fuse the decoration scheme information semantic feature vector and the estimated influencing parameter information association feature vector to obtain an interior decoration semantic association feature matrix; a text convolutional coding unit, used to pass the interior decoration semantic association feature matrix through a text convolutional neural network model as a feature extractor to obtain an engineering material consumption estimation association feature map; a dimensionality reduction unit, used to perform global mean pooling on each feature matrix along the channel dimension of the engineering material consumption estimation association feature map to obtain the engineering material consumption estimation association feature vector; and a collaborative modulation unit, used to perform kernel collaborative modulation based on intrinsic manifold regularization on the engineering material consumption estimation association feature vector to obtain the modulated engineering material consumption estimation association feature vector.

[0046] Specifically, in a specific example of the present application, the fusion matrix unit is used to fuse the semantic feature vector of the decoration scheme information and the associated feature vector of the estimated influencing parameter information to obtain an interior decoration semantic association feature matrix. It should be understood that fusing the semantic feature vector of the decoration scheme information and the associated feature vector of the estimated influencing parameter information can integrate their information to form a more comprehensive and integrated feature representation. This helps improve the feature representation capability, thereby better describing the semantic association characteristics of interior decoration. Fusion of two different types of feature vectors can help establish a semantic association model for interior decoration information. Such an association model can better capture the correlation between decoration scheme information and influencing parameter information, thereby improving the ability to estimate project material consumption. Fusing feature vectors can help extract important information from different types of features, thereby better describing the semantic association characteristics of interior decoration. This helps increase the richness and diversity of features, providing a better data foundation for subsequent feature extraction and modeling. The resulting interior decoration semantic association feature matrix can be used as input to a text convolutional neural network model, helping the model better understand and process the semantic associations of decoration information. This helps improve the accuracy and effectiveness of project material consumption estimation. Fusion of decoration scheme information and influencing parameter information into a semantic association feature matrix can simplify data processing complexity, making the data easier to process and analyze, which helps improve the efficiency and accuracy of data processing.

[0047] Furthermore, the fusion matrix unit is used to: jointly encode the decoration scheme information semantic feature vector and the estimated impact parameter information association feature vector using the following formula to generate the interior decoration semantic association feature matrix;

[0048] Wherein, the formula is:

[0049]

[0050] in represents vector multiplication, M represents the interior decoration semantic association feature matrix, V1 represents the decoration scheme information semantic feature vector, V2 represents the estimated impact parameter information association feature vector, Represents the transpose of the feature vector associated with the estimated impact parameter information.

[0051] Specifically, in one specific example of the present application, the text convolutional encoding unit is configured to pass the interior decoration semantic association feature matrix through a text convolutional neural network model acting as a feature extractor to obtain a correlation feature map for project material consumption estimates. It should be understood that a text convolutional neural network is an effective feature extractor suitable for processing data with spatial relationships, such as natural language text or similar structured data. By inputting the interior decoration semantic association feature matrix into the text convolutional neural network, hidden feature information can be effectively extracted. When processing text data, text convolutional neural networks can learn to represent semantic information, which is particularly important for the semantic association features of interior decoration information. This network model can better understand and model the semantic associations within interior decoration information, thereby improving the accuracy of project material consumption estimates. The text convolutional neural network can learn nonlinear features, thereby better capturing the complex relationships within the interior decoration semantic association feature matrix. This nonlinear modeling can more accurately describe the relationship between decoration information and project material consumption, improving the accuracy of estimates. By inputting the interior decoration semantic association feature matrix into the text convolutional neural network model, an end-to-end learning process can be achieved. Therefore, inputting the interior decoration semantic association feature matrix into such a network model can make the model better adapt to this type of data and improve the prediction accuracy and generalization ability.

[0052] Specifically, in a specific example of the present application, the dimensionality reduction unit is configured to perform global mean pooling on each feature matrix along the channel dimension of the engineering material consumption estimation-related feature map to obtain the engineering material consumption estimation-related feature vector. It should be understood that global mean pooling can effectively compress the information of each feature matrix, converting it into a single feature vector. This helps reduce the feature dimension, lowers computational complexity, and improves model efficiency. Through global mean pooling, feature information on each channel can be integrated into a global feature vector. This helps capture global information from the entire feature map, rather than just local information, thereby improving feature representation and model generalization. Global mean pooling can reduce the number of model parameters and help mitigate the risk of overfitting. By compressing the feature matrix into a single feature vector, the model's complexity can be reduced and its generalization can be improved. Global mean pooling is independent of the spatial position of the feature and therefore has a certain degree of translation invariance. This means that regardless of the feature's position in the image, the feature vector obtained through global mean pooling remains the same, helping to improve the stability and robustness of the model.

[0053] In the process of converting raw data into the final associated feature vector for project material consumption estimates, although various models extract information at different levels, these models may not fully capture all important internal structure of features. For example, when using convolutional neural networks to process the feature vector for decoration scheme information, while CNNs excel at identifying local patterns and features, they are limited in their ability to capture long-range dependencies between features and more complex contextual information. This means that some key internal structure information, such as how specific decorative materials interact to influence overall material consumption, or the deep-seated impact of house area and spatial layout on material demand, may not be fully represented. Furthermore, while concatenating multiple estimated impact parameter information feature vectors to form the associated feature vector for estimated impact parameter information can integrate different features, it is essentially a simple linear combination that fails to account for the inherent connections and complex interactions between features. This can lead to the neglect of some subtle but crucial internal structure information, which in turn affects the accuracy of the final project material consumption estimate. In order to overcome this problem, in a specific example of the present application, the collaborative modulation unit is used to perform kernel collaborative modulation based on intrinsic manifold regularization on the engineering material consumption estimation associated feature vector to obtain the modulated engineering material consumption estimation associated feature vector.

[0054] Specifically, the cooperative modulation unit is used to: first, construct a sub-pixel field correlation matrix of the engineering material consumption estimation correlation feature vector, which is expressed as follows:

[0055]

[0056] Wherein, V represents the estimated associated characteristic vector of the engineering material consumption, v i and v j They represent the eigenvalues of the i-th and j-th positions of the engineering material consumption estimation associated eigenvector, d(v i ,v j ) indicates calculating the Euclidean distance, D i,j Represents the eigenvalue at position (i, j) of the sub-pixel field correlation matrix.

[0057] That is, by introducing the correlation modeling of continuous spatial fields, it is possible to analyze more subtle nonlinear coupling mechanisms between decoration scheme parameters and engineering material consumption. Nonlinear coupling mechanisms may be generated by the microscopic interaction between material physical properties (such as thermal expansion coefficient and stress distribution) and spatial layout, or by dynamic influencing factors such as the material reuse probability and construction process error compensation implicit in the decoration scheme. Through the execution of the sub-pixel field correlation matrix, the multi-dimensional engineering material consumption estimation correlation feature vectors are mapped to the continuous manifold space, and the differential geometric relations in field theory are used to mine the high-order dependency patterns of feature units in the topological structure, thereby constructing an implicit conservation law model that can characterize the dynamic balance between material consumption and parameters, so that the material consumption estimation model has the ability to adaptively analyze complex boundary conditions.

[0058] Secondly, dilated convolution is used to perform kernel feature analysis on the sub-pixel field correlation matrix to obtain the kernel co-activation matrix associated with engineering material consumption estimation, which is expressed as follows:

[0059] M=Conv(D)

[0060] Wherein, D represents the sub-pixel field correlation matrix, Conv represents the convolutional layer, and M represents the engineering material consumption estimation correlation core domain synergistic activation matrix.

[0061] That is, by introducing a convolution kernel with an exponential expansion factor, the system can establish a multi-resolution perception mechanism along the spatial dimension while maintaining parameter efficiency, thereby parsing the chain reaction pattern hidden in the material consumption estimation. In essence, it is to deconstruct the domains with different physical meanings in the sub-pixel field correlation matrix, and establish a dynamic balance equation between the material category consumption and the spatial topological constraints through the kernel domain co-activation mechanism. Specifically, the multi-level jump sampling characteristics of the dilated convolution enable the model to simultaneously capture the synergistic effects of short-range strong correlation and long-range weak correlation in the material consumption estimation, thereby generating the engineering material consumption estimation-related kernel domain co-activation matrix, and forming a multi-scale modeling capability for the chain response of material consumption in complex decoration projects.

[0062] Then, the sub-pixel field correlation matrix is decoupled in the spectral domain to extract a set of hierarchical intrinsic component coding vectors associated with engineering material consumption estimation, which can be expressed as follows:

[0063]

[0064] Where T represents the transpose of the vector, Λ represents the diagonal matrix, λ1 and λ m They represent the first and mth eigenvalues of the diagonal matrix respectively, U represents the set of hierarchical eigencomponent coding vectors associated with engineering material consumption estimation, x1, x2, x mThey represent the first, second and mth engineering material consumption estimation associated hierarchical eigencomponent coding vectors respectively.

[0065] Specifically, through a hierarchical screening mechanism of quantized intrinsic components, the material consumption modes corresponding to different energy levels in the sub-pixel field correlation matrix are separated, thereby constructing a multi-resolution explanatory model for material consumption estimation. Specifically, through the orthogonal completeness of the layered intrinsic components, a set of layered intrinsic component encoding vectors associated with engineering material consumption estimation is generated. This enables the system to accurately analyze potential material consumption critical points in decoration projects, and the estimation model has the ability to decouple and reconstruct nonlinear correlation patterns.

[0066] Next, the set of hierarchical intrinsic component encoding vectors associated with the engineering material consumption estimate is input into the converter model based on the self-attention mechanism to obtain a set of adaptive intrinsic component encoding vectors associated with the engineering material consumption estimate, which is expressed as follows:

[0067] Y=Transformer{[x1,x2,…,x m ]}=[y1,y2,…,y m ]

[0068] Among them, Transformer represents the converter model based on the self-attention mechanism, Y represents the set of engineering material consumption estimation associated adaptation intrinsic component encoding vectors, y1, y2, y m They represent the first, second and mth engineering material consumption estimation associated adaptation eigencomponent coding vectors respectively.

[0069] Specifically, the unique cross-scale context-awareness of the self-attention mechanism is leveraged to construct an asymmetric dependency network between the encoding vectors of the hierarchical intrinsic components associated with project material consumption estimation, enabling self-consistent adaptation of the material consumption estimation model to dynamic construction scenarios. Specifically, attention weights are used to quantize and reorganize the set of encoding vectors of the hierarchical intrinsic components associated with project material consumption estimation, thereby improving the generalization of estimates in multivariate coupling scenarios while reducing reliance on manual experience.

[0070] Then, each engineering material consumption estimate associated adaptive intrinsic component coding vector in the set of engineering material consumption estimate associated adaptive intrinsic component coding vectors is bilinearly interacted with the engineering material consumption estimate associated core domain collaborative activation matrix to obtain a set of engineering material consumption estimate associated intrinsic component core collaborative coding vectors, which is expressed as follows:

[0071]

[0072] in, represents matrix multiplication, S represents the characteristic scale of the synergistic activation matrix of the engineering material consumption estimation correlation core domain, and y i represents the i-th engineering material consumption estimation associated adaptation intrinsic component coding vector, L represents the length of the engineering material consumption estimation associated adaptation intrinsic component coding vector, z i Represents the kernel collaborative coding vector of the intrinsic component associated with the i-th engineering material consumption estimate.

[0073] Specifically, bilinear tensor operations are used to achieve the cross-fusion of the heterogeneous feature spaces of the adaptive intrinsic component coding vectors associated with engineering material consumption estimation and the collaborative activation matrix of the core domain associated with engineering material consumption estimation. This constructs a composite feature representation that both retains the inherent laws of material consumption calculation and reflects the dynamic coupling relationship between actual engineering parameters. This is the set of collaborative coding vectors of the intrinsic component kernels associated with engineering material consumption estimation. This interactive mechanism enables the system to automatically mine complex dependencies that are difficult to quantify using traditional manual estimation, such as the implicit associations between spatial layout and material categories, and the nonlinear constraints between area parameters and construction processes. This forms a dynamic probability field for material consumption in the feature space, capturing the conditional sensitivity of different decoration parameter combinations to material consumption through a bilinear mapping matrix. This allows the system to adaptively adjust the estimation weights based on the strength of feature interactions when faced with complex scenarios such as irregular spaces and non-standard construction plans, avoiding the estimation bias caused by local parameter mutations caused by traditional linear models and improving robustness under complex engineering conditions.

[0074] Finally, the set of the engineering material consumption estimation-related intrinsic component kernel collaborative coding vectors is chain-fused to obtain the modulated engineering material consumption estimation-related feature vector, which is expressed as follows:

[0075] V'=Concat{z1,z2,…,z m}

[0076] Among them, Concat represents the cascade function, z1, z2, z m They represent the first, second and mth engineering material consumption estimation associated eigencomponent kernel collaborative coding vectors respectively, and V' represents the engineering material consumption estimation associated eigenvector after modulation.

[0077] Specifically, a sequential cascade fusion strategy constructs a feature structure with process topology awareness, enabling material consumption estimates to reflect the parameter transfer characteristics of process transitions in real-world construction. This generates a modulated, causal feature vector for project material consumption estimates. This vector uses a chain-based weight transfer mechanism to model the material consumption transmission coefficient between construction phases. This allows the system to identify, for example, the potential impact of an abnormal pre-buried conduit during the water and electricity renovation phase on subsequent wall leveling material consumption, thereby improving the overall consistency of material consumption estimates in multi-process coupled scenarios.

[0078] In this embodiment of the present application, the interior decoration information analysis module 140 is configured to pass the modulated feature vector associated with the estimated construction material consumption through a decoder to obtain a decoded value. This decoded value represents the final estimated construction material consumption for the target interior decoration project. It should be understood that the decoder helps map the modulated feature vector back to the original construction material consumption space. The decoder can recover the information related to construction material consumption hidden in the modulated feature vector, thereby achieving information restoration and reconstruction. The decoded value, as the output of the model, directly represents the model's estimated construction material consumption. These decoded values are the final prediction results and can be directly used to formulate material procurement plans and budget arrangements for the decoration project. The decoder helps convert the modulated feature vector into a more concrete and interpretable form, thereby translating the prediction problem into actual construction material consumption. The decoded value provides an intuitive and easy-to-understand quantitative indicator that can be used to evaluate and plan the material consumption required for decoration projects. The decoded value generated by the decoder helps explain how the model derives the estimated construction material consumption. This interpretability is crucial for engineers and designers to understand the model's decision-making process and the reasons behind its results. The decoder can map the learned high-level abstract features back to the original data space, thus achieving the reverse mapping from feature space to original data space. This mapping helps to convert the modulated features learned by the model into actual engineering material consumption estimates.

[0079] In summary, the data acquisition and analysis system and method for interior decoration based on the embodiment of the present application extracts the characteristics of the decoration scheme information and the estimated influencing parameter information of the target interior decoration project, constructs and decodes the associated feature map, so as to achieve accurate estimation of the material consumption of the interior decoration project, thereby effectively solving the current problems of low accuracy, waste of materials or delay in construction period existing in the single means based on manual estimation.

[0080] As described above, the data acquisition and analysis system 100 for interior decoration according to the embodiments of the present application can be implemented in various terminal devices, such as a server of the data acquisition and analysis system for interior decoration. In one example, the data acquisition and analysis system 100 for interior decoration can be integrated into the terminal device as a software module and / or a hardware module. For example, the data acquisition and analysis system 100 for interior decoration can be a software module in the operating system of the terminal device, or can be an application developed for the terminal device; of course, the data acquisition and analysis system 100 for interior decoration can also be one of the many hardware modules of the terminal device.

[0081] Alternatively, in another example, the data acquisition and analysis system 100 for interior decoration and the terminal device may also be separate devices, and the data acquisition and analysis system 100 for interior decoration may be connected to the terminal device via a wired and / or wireless network and transmit interactive information in accordance with an agreed data format.

[0082] Figure 5 Flowchart of the data collection and analysis method for interior decoration according to an embodiment of the present application. Figure 5 As shown, the data collection and analysis method for interior decoration according to the embodiment of the present application includes the steps of: S110, obtaining decoration scheme information and estimated impact parameter information of the target interior decoration project, wherein the decoration scheme information of the target interior decoration project includes the engineering material category information required for the corresponding decoration scheme, and the estimated impact parameter information of the target interior decoration project includes the house area and space layout; S120, extracting the decoration scheme information semantic feature vector and the estimated impact parameter information association feature vector from the decoration scheme information of the target interior decoration project and the estimated impact parameter information of the target interior decoration project respectively; S130, constructing an engineering material consumption estimated association feature vector between the decoration scheme information semantic feature vector and the estimated impact parameter information association feature vector, and performing kernel co-modulation based on intrinsic manifold regularization on the engineering material consumption estimated association feature vector to obtain a modulated engineering material consumption estimated association feature vector; S140, passing the modulated engineering material consumption estimated association feature vector through a decoder to obtain a decoded value, and the decoded value is used to represent the final engineering material consumption estimated amount of the target interior decoration project.

[0083] Here, those skilled in the art will appreciate that the specific operations of each step in the above-mentioned data collection and analysis method for interior decoration have been described in detail in the above reference. Figures 1 to 4 The data acquisition and analysis system for interior decoration has been described in detail, and therefore, its repeated description will be omitted.

[0084] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present invention.

[0085] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0086] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0087] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0088] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0089] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0090] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present invention should be included in the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. A data acquisition and analysis system for interior decoration, characterized in that: include: An interior decoration information collection module, configured to obtain decoration scheme information and estimated impact parameter information of a target interior decoration project, wherein the decoration scheme information of the target interior decoration project includes information on the categories of engineering materials required for the corresponding decoration scheme, and the estimated impact parameter information of the target interior decoration project includes the house area and spatial layout; an interior decoration information processing module, configured to extract a decoration scheme information semantic feature vector and an estimated impact parameter information association feature vector from the decoration scheme information of the target interior decoration project and the estimated impact parameter information of the target interior decoration project, respectively; an interior decoration information fusion module, configured to construct an estimated engineering material consumption correlation feature vector between the decoration scheme information semantic feature vector and the estimated influencing parameter information correlation feature vector, and perform kernel co-modulation based on intrinsic manifold regularization on the estimated engineering material consumption correlation feature vector to obtain a modulated estimated engineering material consumption correlation feature vector; The interior decoration information analysis module is used to pass the modulated engineering material consumption estimate associated feature vector through a decoder to obtain a decoded value, and the decoded value is used to represent the final engineering material consumption estimate of the target interior decoration project.

2. The data acquisition and analysis system for interior decoration according to claim 1, characterized in that: The interior decoration information processing module includes: A decoration scheme information extraction unit, configured to perform convolution coding on the decoration scheme information of the target interior decoration project to obtain a semantic feature vector of the decoration scheme information; The estimated impact parameter information extraction unit is used to segment the estimated impact parameter information of the target interior decoration project and then encode it to obtain the estimated impact parameter information associated feature vector.

3. The data acquisition and analysis system for interior decoration according to claim 2, characterized in that: The decoration scheme information extraction unit includes: a semantic understanding subunit, configured to obtain a decoration scheme information feature vector from the decoration scheme information of the target interior decoration project using a semantic understanding model; The first convolutional encoding subunit is used to pass the decoration scheme information feature vector through an information extractor based on a first convolutional neural network model to obtain a decoration scheme information semantic feature vector.

4. The data acquisition and analysis system for interior decoration according to claim 3, characterized in that: The semantic understanding subunit includes: a word vector conversion secondary sub-unit, configured to use the word embedding layer of the semantic understanding model to map each word in the decoration scheme information of the target interior decoration project into a word vector to obtain a decoration scheme information word vector sequence; a semantic processing secondary sub-unit, configured to process the decoration scheme information word vector sequence using the Bert model of the semantic understanding model to obtain a decoration scheme information word feature vector sequence; The context encoding secondary sub-unit is used to use the bidirectional LSTM network of the semantic understanding model to context encode the decoration scheme information word feature vector sequence to obtain the weather information feature vector.

5. The data acquisition and analysis system for interior decoration according to claim 4, characterized in that: The estimated impact parameter information extraction unit includes: a context encoding subunit, configured to perform word segmentation processing on the estimated impact parameter information of the target interior decoration project and then pass it through a first context encoder based on a converter to obtain a plurality of estimated impact parameter information feature vectors; The vector cascading subunit is configured to cascade the plurality of estimated influencing parameter information feature vectors to obtain an estimated influencing parameter information associated feature vector.

6. The data acquisition and analysis system for interior decoration according to claim 5, characterized in that: The interior decoration information fusion module includes: a fusion matrix unit, configured to fuse the decoration scheme information semantic feature vector and the estimated influencing parameter information association feature vector to obtain an interior decoration semantic association feature matrix; A text convolutional coding unit, configured to pass the interior decoration semantic association feature matrix through a text convolutional neural network model as a feature extractor to obtain an engineering material consumption estimation association feature map; A dimensionality reduction unit, configured to perform global mean pooling on each feature matrix along the channel dimension of the engineering material consumption estimation correlation feature map to obtain the engineering material consumption estimation correlation feature vector; The collaborative modulation unit is used to perform kernel collaborative modulation based on intrinsic manifold regularization on the engineering material consumption estimation associated feature vector to obtain the modulated engineering material consumption estimation associated feature vector.

7. The data acquisition and analysis system for interior decoration according to claim 6, characterized in that: The fusion matrix unit is used to: jointly encode the decoration scheme information semantic feature vector and the estimated impact parameter information association feature vector according to the following formula to generate the interior decoration semantic association feature matrix; Wherein, the formula is: in represents vector multiplication, M represents the interior decoration semantic association feature matrix, V1 represents the decoration scheme information semantic feature vector, V2 represents the estimated impact parameter information association feature vector, Represents the transpose of the feature vector associated with the estimated impact parameter information.

8. The data acquisition and analysis system for interior decoration according to claim 7, characterized in that: The cooperative modulation unit is configured to: Constructing a sub-pixel field correlation matrix of the engineering material consumption estimation correlation feature vector; Using dilated convolution to perform kernel domain feature analysis on the sub-pixel field correlation matrix to obtain a kernel domain synergistic activation matrix associated with engineering material consumption estimation; Performing spectral domain decoupling on the sub-pixel field correlation matrix to extract a set of hierarchical intrinsic component coding vectors associated with engineering material consumption estimation; Inputting the set of hierarchical intrinsic component encoding vectors associated with the engineering material consumption estimate into a converter model based on a self-attention mechanism to obtain a set of adapted intrinsic component encoding vectors associated with the engineering material consumption estimate; Performing bilinear interaction between each engineering material consumption estimate associated adaptive intrinsic component coding vector in the set of engineering material consumption estimate associated adaptive intrinsic component coding vectors and the engineering material consumption estimate associated core domain collaborative activation matrix to obtain a set of engineering material consumption estimate associated intrinsic component core collaborative coding vectors; The set of engineering material consumption estimation-related intrinsic component kernel collaborative coding vectors is chain-fused to obtain the modulated engineering material consumption estimation-related feature vector.

9. A data collection and analysis method for interior decoration, characterized in that: include: Obtaining decoration scheme information and estimated impact parameter information of a target interior decoration project, wherein the decoration scheme information of the target interior decoration project includes information on the categories of engineering materials required for the corresponding decoration scheme, and the estimated impact parameter information of the target interior decoration project includes house area and space layout; extracting a decoration scheme information semantic feature vector and an estimated impact parameter information association feature vector from the decoration scheme information of the target interior decoration project and the estimated impact parameter information of the target interior decoration project respectively; Constructing an estimated engineering material consumption correlation feature vector between the decoration scheme information semantic feature vector and the estimated influencing parameter information correlation feature vector, and performing kernel co-modulation based on intrinsic manifold regularization on the estimated engineering material consumption correlation feature vector to obtain a modulated estimated engineering material consumption correlation feature vector; The modulated engineering material consumption estimate associated feature vector is passed through a decoder to obtain a decoded value, and the decoded value is used to represent the final engineering material consumption estimate of the target interior decoration project.

Images