Cloud Architecture Clustering Management Method for Big Data of Pet Allergen Detection Results

Abstract

The present invention relates to the technical field of data processing, and specifically discloses a cloud architecture clustering management method for big data of pet allergen detection results, including a collection step, a generalized function set construction step, a neutral virtual pet model construction step, a clustering step, and an output step; by collecting pet allergen detection data and constructing a pet allergy response function, a generalized function set is formed; then, using the neutral virtual pet function set as a benchmark, single-value characterization and clustering analysis are performed on the distances between different function sets to obtain a clustering interval and an allergy determination result concentration; finally, the pet allergy determination result and the corresponding confidence level are output; this solution can improve the accuracy of the assessment of pet allergy risks, has both flexibility and scalability, is applicable to large-scale pet allergy detection and data management scenarios, and data sharing can be achieved through a unified cloud architecture.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and particularly to a cloud architecture clustering management method for big data of pet allergen detection results. Background Art

[0002] In recent years, with the improvement of people's living standards and the popularization of the trend of keeping pets, pets have gradually become an important part of family members. At the same time, the demand for pet health management is also increasing day by day. Among them, pet allergen detection, as an important diagnostic method, is widely used in pet hospitals, research institutions and related testing centers. Pet allergic reactions are usually caused by environmental factors (such as pollen, dust mites, molds, etc.) or food ingredients (such as beef, chicken, grains, etc.), and are manifested as symptoms such as skin itching, ear inflammation, gastrointestinal discomfort, respiratory problems, etc. Therefore, accurately identifying the allergens of individual pets is of great significance for their health management and the improvement of their quality of life.

[0003] At present, pet allergen detection methods mainly include serum IgE antibody detection, skin prick test (SPT), food elimination test and molecular diagnostic techniques, etc. Among them, serum IgE detection is a relatively common method, which judges the sensitivity of a pet to a specific allergen by analyzing the level of specific immunoglobulins in the pet's blood. However, due to the differences in pet species, genetic backgrounds, living environments and feeding methods, the occurrence mechanism of allergic reactions is relatively complex, and traditional detection methods still face many challenges in data analysis, information integration and personalized application. With the development of big data and artificial intelligence technologies, how to effectively manage, analyze and utilize pet allergen detection results to make them more clinically valuable has become an urgent problem to be solved in the current industry.

[0004] Currently, pet allergen detection still faces the following main problems:

[0005] 1. Difficulties in data storage and sharing: Current pet allergen detection data is usually stored in independent databases of different laboratories or pet hospitals. The data formats and storage methods lack unified standards, resulting in difficulties in data integration. At the same time, due to the limitations of privacy protection and technical barriers, cross-institutional data sharing and interoperability are poor, and a complete allergen database cannot be formed, thus affecting the overall utilization value of the data.

[0006] 2. Limited data analysis capabilities: Traditional analysis methods mainly rely on linear statistics or simple rule matching, and it is difficult to fully explore the correlation relationships between multi-dimensional data and accurately identify the allergy patterns of different pet groups. For example, certain specific breeds of pets may be more prone to react to certain allergens, and this information is difficult to effectively extract through traditional methods.

[0007] 3. Intelligent diagnosis lacking big data support: The occurrence of pet allergic reactions is affected by multiple factors, such as genetic background, breeding environment, climate change, etc. Existing detection methods usually only focus on individual data and fail to conduct intelligent analysis by integrating large-scale data, making it difficult to achieve accurate prediction of allergy risks and diagnostic and treatment recommendations.

[0008] 4. Insufficient real-time update and feedback mechanism: Current detection reports are usually presented in the form of static documents, lacking the ability of dynamic update and trend prediction. For example, the reactions of some pets to specific allergens may vary with age, immune status, or environmental changes, but the existing system cannot achieve long-term data tracking, resulting in the lack of scientific basis and real-time adjustment ability for allergy management strategies.

[0009] To address the above problems, it is urgent to construct a big data clustering management method based on cloud architecture, integrate pet allergen detection data into the cloud, and classify, cluster, and analyze trends in the data through machine learning algorithms to achieve efficient storage, intelligent analysis, and real-time feedback of detection data, thereby providing a more accurate, scientific, and personalized solution for pet allergy management. Summary of the Invention

[0010] To address the above problems, the object of the present invention is to propose a cloud architecture clustering management method for big data of pet allergen detection results, including the following steps:

[0011] S1. Collection step: Collect pet allergen detection data to obtain an allergen detection data set;

[0012] S2. Construction of a generalized function set step: Construct a pet allergy response function, and form a generalized function set by combining the pet allergy response function with the allergen detection data set;

[0013] S3. Construction of a neutral virtual pet model step: Referring to the form of the generalized function set, construct a neutral virtual pet model, and the neutral virtual pet model has a neutral virtual pet function set;

[0014] S4. Clustering step: Perform single-value characterization and clustering analysis on the distances of different generalized function sets relative to the neutral virtual pet function set to obtain the generalized function distance, clustering interval, and allergy determination result concentration;

[0015] S5. Output step: Based on the generalized function distance, clustering interval, and allergy determination result concentration, output the pet allergy determination result and confidence level.

[0016] Further, in step S1, the allergen detection data set is formally expressed as , where is the pet serial number, , is the allergen detection data set Dimension; For the th pet's allergen detection data.

[0017] Furthermore, in step S2, the pet allergy response function is formally expressed as , and the generalized function set is formally expressed as , satisfying:

[0018]

[0019]

[0020]

[0021]

[0022] Wherein, a and b are the reaction lower threshold and the reaction upper threshold respectively, both of which are positive real numbers and satisfy a < b; And Are respectively the th pet's allergen detection data's th data component's detection value and weight value; Is the pet allergy response function; Is the allergy determination result.

[0023] Furthermore, step S2 also includes: introducing a default allergen detection data set , and the default allergen detection data set As the first set of data sets in the generalized function set , constructing an initial generalized function set With an initial weight value ; The initial generalized function set Satisfies:

[0024]

[0025] Wherein, the default allergen detection data set Refers to the set of allergen detection data when a certain pet has no detection results or data missing on some allergens; , And Respectively represent the allergen detection data, the pet allergy response function and the allergy determination result when a certain pet has no detection results or data missing on some allergens;

[0026] Satisfies:

[0027] T1: Assume that a pet is first exposed to a certain allergen and has no allergic reaction in the historical record pet, then ;

[0028] T2: Assume that the pet in the historical record was first exposed to a certain allergen and had an allergic reaction, then ;

[0029] T3: Assume that the pet in the historical record was exposed to a certain allergen multiple times and had both allergic reactions and non-allergic reactions, then , where is the number of exposures, is the number of allergic reactions.

[0030] Further, step S3 is specifically as follows: Refer to the form of the generalized function set to construct a neutral virtual pet function set with the neutral virtual pet function set .

[0031] Further, in step S3, refer to the form of the generalized function set to establish a neutral virtual pet function set , satisfying:

[0032]

[0033] where represents a neutral virtual pet with no allergic reaction to all participating test items, is a neutral assignment function, indicating that the neutral virtual pet assigns all participating test items to the reaction lower threshold.

[0034] Further, step S4 specifically includes:

[0035] S41: Perform a single-value characterization of the distances between different generalized function sets and the neutral virtual pet function set to obtain the generalized function distance;

[0036] S42: Conduct a cluster analysis on the generalized function distance to obtain a cluster interval, and determine the allergic determination result concentration based on the cluster interval.

[0037] Further, step S41 specifically includes:

[0038] Introduce a distance function to perform a single-value characterization of the distances between different generalized function sets and the neutral virtual pet function set. Let any generalized function set and the neutral virtual pet function set have a generalized function distance satisfying:

[0039]

[0040] where and respectively represent the values of the generalized function set and the neutral virtual pet function set on the th feature dimension.

[0041] Furthermore, step S42 specifically includes:

[0042] S421. Perform cluster analysis on the generalized function distance represented by a single value to obtain a set of cluster groups , satisfying:

[0043]

[0044] Among them, is the serial number of the cluster group, represents the cluster group with the serial number , is the total number of cluster groups;

[0045] S422. For different cluster groups, capture the allergy determination results of the elements within the group, and calculate the concentration of the allergy determination results, satisfying:

[0046]

[0047] Among them, , and respectively represent the number of allergic reactions and severe allergic reactions, the total number of elements within the cluster group, and the concentration of the allergy determination results of the cluster group with the serial number .

[0048] Beneficial effects

[0049] The present invention proposes a cloud architecture clustering management method for big data of pet allergen detection results. Compared with the prior art, it has the following beneficial effects:

[0050] 1. Clearer baseline reference: By constructing a "neutral virtual pet model" and endowing the neutral virtual pet function set as a unified baseline or reference object, the allergen detection data of different pets can be compared and cluster analyzed in the same reference system, thus greatly reducing the deviation of detection results caused by individual differences of pets.

[0051] 2. More accurate cluster analysis: By calculating the distance between any generalized function set and the neutral virtual pet function set, and performing single-value characterization and cluster analysis on this distance, not only can the allergy similarity and difference degrees between different pets be effectively discovered, but also the clustering interval can be further refined, and accurate classification can be carried out in combination with the concentration of allergy determination results (such as mild allergy, severe allergic reaction, etc.), thereby improving the discrimination and accuracy of allergen detection results.

[0052] 3. Improvement in big data processing efficiency: The present invention uniformly manages pet allergen detection data under a cloud architecture. By integrating multi-dimensional information such as detection values, weight values, and allergy determination results in the form of a "generalized function set", it not only facilitates efficient calculation and query under a large data scale but also enables quick retrieval for subsequent historical data traceability, statistical analysis, and result prediction, thus enhancing the overall processing efficiency.

[0053] 4. Wide range of applicable scenarios: Due to the introduction of mechanisms such as a default allergen detection data set and an initial generalized function set, it can be compatible with data loss or incomplete initial data in various scenarios; and by statistically analyzing multiple exposures to the same allergen (the ratio of the number of exposures to the number of allergic reactions), comprehensively evaluating the sensitivity of the pet, it greatly enhances the robustness and applicability of the method. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 It is a schematic flowchart of the method of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0055] To deepen the understanding of the present invention, the following will further elaborate on the present invention in conjunction with embodiments. These embodiments are only used to explain the present invention and do not constitute a limitation to the protection scope of the present invention.

[0056] Embodiment 1

[0057] According to Figure 1 as shown, this embodiment provides a cloud architecture clustering management method for big data of pet allergen detection results, including the following steps:

[0058] S1. Collection step: Collect pet allergen detection data to obtain an allergen detection data set;

[0059] S2. Construction of the generalized function set step: Construct a pet allergy response function, and form a generalized function set by combining the pet allergy response function with the allergen detection data set;

[0060] S3. Construction of a neutral virtual pet model step: Referring to the form of the generalized function set, construct a neutral virtual pet model, and the neutral virtual pet model has a neutral virtual pet function set;

[0061] S4. Clustering step: Perform single-value characterization and clustering analysis on the distances of different generalized function sets relative to the neutral virtual pet function set to obtain the generalized function distance, clustering interval, and allergy determination result concentration;

[0062] S5. Output step: Based on the generalized function distance, clustering interval, and allergy determination result concentration, output the pet allergy determination result and confidence level.

[0063] Further, in step S1, the allergen detection data set is formally represented as , where is the pet serial number, , is the dimension of the allergen detection data set ; is the th pet's allergen detection data.

[0064] Further, in step S2, the pet allergy response function is formally represented as , and the generalized function set is formally represented as , satisfying:

[0065]

[0066]

[0067]

[0068]

[0069] Among them, a and b are the reaction lower threshold and the reaction upper threshold respectively, both of which are positive real numbers and satisfy a < b; and are the detection value and the weight value of the th item data component of the allergen detection data of the th pet respectively; is the pet allergy response function; is the allergy determination result.

[0070] Further, step S2 also includes: introducing a default allergen detection data set , and the default allergen detection data set is used as the first group of data sets in the generalized function set to construct an initial generalized function set with an initial weight value ; The initial generalized function set satisfies:

[0071]

[0072] Among them, the default allergen detection data set refers to the set of allergen detection data when there is no detection result or data missing for a certain pet on some allergens; , and respectively represent the allergen detection data, the pet allergy response function, and the allergy determination result when there is no detection result or data missing for a certain pet on some allergens;

[0073] Satisfy:

[0074] T1: Assume that the pet in the historical record is first exposed to a certain allergen and has no allergic reaction, then ;

[0075] T2: Assume that the pet in the historical record is first exposed to a certain allergen and has an allergic reaction, then ;

[0076] T3: Assume that the pet in the historical record is exposed to a certain allergen multiple times and there are situations of having allergic reactions and no allergic reactions, then , where is the number of exposures, is the number of times of having allergic reactions.

[0077] Furthermore, step S3 is specifically as follows: Refer to the form of the generalized function set to construct a neutral virtual pet function set with the neutral virtual pet function set .

[0078] Furthermore, in step S3, refer to the form of the generalized function set to establish a neutral virtual pet function set , satisfying:

[0079]

[0080] Among them, represents a neutral virtual pet that has no allergic reaction to all test items participated in, is a neutral assignment function, indicating that the neutral virtual pet assigns all test items participated in to the reaction lower threshold.

[0081] Furthermore, step S4 specifically includes:

[0082] S41. Make a single-value characterization of the distance between different generalized function sets and the neutral virtual pet function set to obtain the generalized function distance;

[0083] S42. Conduct a cluster analysis on the generalized function distance to obtain a cluster interval, and determine the allergy determination result concentration based on the cluster interval.

[0084] Furthermore, step S41 specifically includes:

[0085] Introduce a distance function to make a single-value characterization of the distance between different generalized function sets and the neutral virtual pet function set. Let any generalized function set With the neutral virtual pet function set The generalized function distance Satisfies:

[0086]

[0087] Wherein, And Respectively represent the values of the generalized function set and the neutral virtual pet function set on the th feature dimension.

[0088] Furthermore, step S42 specifically includes:

[0089] S421. Perform clustering analysis on the generalized function distance represented by a single value To obtain a set of clustering groups Satisfies:

[0090]

[0091] Wherein, Is the clustering group serial number, Represents the clustering group with the serial number Is the total number of clustering groups;

[0092] S422. For different clustering groups, capture the allergy determination results of the elements within the group And calculate the concentration of the allergy determination results, satisfying:

[0093]

[0094] Wherein, , And Respectively represent the number of allergic reactions and severe allergic reactions, the total number of elements within the clustering group, and the concentration of the allergy determination results of the clustering group with the serial number

[0095] Example 2

[0096] Based on the distributed cloud platform of the pet medical center, this example uniformly manages and performs clustering analysis on the allergen detection data of pets from different regions and different species to evaluate the allergy risk level of pets. Its implementation process and algorithm description are as follows:

[0097] 1. Data collection and preprocessing

[0098] 1.1 Data collection:

[0099] Each pet hospital or detection point uploads the detection results to the cloud server to form an original data set .

[0100] wherein is the pet serial number, , is the allergen detection data set dimension; is the allergen detection data of the th pet, including multiple dimensions or eigenvalue , . Each dimension can correspond to a specific allergen (such as cats to dust mites, cat scratch ringworm, pollen, etc.), or can correspond to different physiological indicators.

[0101] 1.2 Data cleaning and default processing

[0102] If a certain pet has no test result or data missing on some allergens, it can be set to , and initialize the corresponding weight according to the historical exposure record (whether there has been an allergy) ;

[0103] For repeated or abnormal values, threshold rules or statistical methods (such as extreme value elimination) can be set for pre-cleaning;

[0104] Finally, a relatively complete and usable data set is formed and ready to enter the subsequent general function set construction link.

[0105] 2. General function set construction

[0106] 2.1 Pet allergy response function

[0107] To measure the comprehensive sensitivity of each pet to multiple allergens, a pet allergy response function is introduced, satisfying:

[0108]

[0109]

[0110] wherein is the test value of the pet on the kth allergen, is the corresponding weight, indicating the importance of this dimension in the comprehensive sensitivity; the sum of the weights is 1 for normalization.

[0111] 2.2 Allergy determination result

[0112] The allergy situation is determined by the following formula:

[0113]

[0114] where a and b are the reaction lower threshold and reaction upper threshold respectively, which can be determined according to large-scale statistics or clinical experience.

[0115] 2.3 Formation of the generalized function set

[0116] For each pet, unify and encapsulate multiple pieces of information such as to form a generalized function set; this generalized function set contains both the original detection data and multi-dimensional elements such as the comprehensive sensitivity and the determination result, facilitating subsequent unified analysis.

[0117] 3. Creation of the neutral virtual pet model

[0118] 3.1 Neutral pet hypothesis

[0119] In actual detection, the characteristic differences of different pets often lead to large differences in the response values to the same allergen. To reduce the influence brought by individual differences, this embodiment constructs a "neutral virtual pet" and assumes that it is in a state of "lower threshold" for all detection items and has no allergic reaction.

[0120] 3.2 Function set of the neutral virtual pet

[0121] Referring to the form of define:

[0122]

[0123] Among them, represents that the detection values of each dimension are assigned as (or a smaller neutral value), can take a fixed constant so that the model is always in a "non-allergic" state, that is, no allergic reaction; this function set is used to compare with the function set of real pets and provide a unified reference benchmark.

[0124] 4. Calculation of the generalized function distance and clustering analysis

[0125] 4.1 Distance calculation (single-value characterization)

[0126] To measure the difference degree between the real pet and the neutral model, a distance function is introduced:

[0127]

[0128] This formula can be regarded as the weighted Euclidean distance of multi-dimensional detection values. On the one hand, it comprehensively considers the data of different allergens, and on the other hand, it can be standardized (such as dividing by ) to control the distance magnitude.

[0129] Implementation of the clustering algorithm:

[0130] Data preparation: The distance values corresponding to all pets Collect into a set of single-value sequences;

[0131] Clustering method: It can adopt , hierarchical clustering ( ), or distance-threshold-based clustering (such as ), etc.;

[0132] Process example: If is adopted, regard as one-dimensional eigenvalue, set the number of clusters, iteratively update the cluster centers and assign them to the nearest cluster until convergence, and finally obtain the set of clustering groups.

[0133] 4.2 Concentration of allergic determination results

[0134] After obtaining each clustering group, count the distribution of allergic reaction types of the elements within the group:

[0135] Calculate the proportions of severe allergic reactions and general allergic reactions, that is:

[0136]

[0137] Among them, is the total number of pets within the group, is the number of pets with allergic reactions (including general and severe allergic reactions); thus, the concentration or risk level of this clustering group in the "allergy" dimension can be quickly evaluated, providing valuable reference for clinical diagnosis and management.

[0138] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments. What is described in the above embodiments and the specification only illustrates the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements all fall within the scope of the present invention claimed. The scope of the present invention claimed is defined by the appended claims and their equivalents.

Claims

1. A cloud architecture clustering management method for big data of pet allergen detection results, characterized by: The following steps are involved: S1. Collection step: Collect pet allergen detection data to obtain an allergen detection dataset; the allergen detection dataset is formally represented as ,in The pet serial number. , Allergen detection dataset Dimensions; For the Allergen testing data of pets; S2. Step of constructing a generalized function set: construct a pet allergy response function, and combine the pet allergy response function and the allergen detection data set to form a generalized function set; the pet allergy response function is formally expressed as , the generalized function set is formally expressed as ,satisfy: ； ； Among them, a and b are the lower threshold and upper threshold of the reaction, respectively, both are positive real numbers, and satisfy a <b; and Respectively Allergen testing data for pets The detection value and weight value of the data component; is the pet allergy response function; Allergy determination result; S3, step of constructing a neutral virtual pet model: constructing a neutral virtual pet model with reference to the form of a generalized function set, wherein the neutral virtual pet model has a neutral virtual pet function set; ,satisfy: ； in, Indicates a neutral virtual pet with no allergic reaction to all the test items. is a neutral assignment function, which means that all the detection items in which the neutral virtual pet participates are assigned the lower threshold of the response S4, clustering step: performing single value characterization and cluster analysis on the distances of different generalized function sets relative to the neutral virtual pet function set to obtain the generalized function distance, clustering interval and allergy determination result concentration; step S4 specifically includes: S41, performing a single-value characterization on the distances of different generalized function sets relative to the neutral virtual pet function set to obtain a generalized function distance; specifically comprising: Introduce the distance function to characterize the distance between different generalized function sets and the neutral virtual pet function set. With neutral virtual pet feature set The generalized functional distance satisfy: ； in, and denote the generalized function set and the neutral virtual pet function set in the The value of the feature dimension; S42, performing cluster analysis on the generalized functional distance to obtain a cluster interval, and determining the concentration of the allergy determination result based on the cluster interval; specifically including: S421. Generalized functional distance using single value representation Perform cluster analysis to obtain a set of cluster groups ,satisfy: ； in, is the cluster group number, Indicates the serial number is The clustering group, is the total number of cluster groups; S422. For different cluster groups, capture the allergy determination results of the elements in the group , calculate the concentration of allergy determination result, satisfying: ； in, , and Respectively indicate the serial number The number of allergic reactions and severe allergic reactions in the cluster group, the total number of elements in the cluster group, and the concentration of allergy judgment results; S5. Output step: output the pet allergy determination result and confidence level based on the generalized functional distance, cluster interval and allergy determination result concentration.

2. The cloud architecture clustering management method for pet allergen detection result big data according to claim 1, characterized in that: Step S2 also includes: introducing a default allergen detection data set , the default allergen detection dataset As a generalized function set The first set of data sets in constructs with initial weight values The initial generalized function set ; Initial generalized function set satisfy: ； Among them, the default allergen detection dataset It refers to the collection of allergen test data of a certain pet when there is no test result or data missing on some allergens; , and They respectively represent the allergen detection data, pet allergy response function and allergy determination results of a certain pet when there is no detection result or data missing on some allergens; satisfy: T1: Assuming that the pet with a history record is exposed to a certain allergen for the first time and has no allergic reaction, then ; T2: Assuming that the pet with a historical record is exposed to a certain allergen for the first time and has an allergic reaction, then ; T3: Assuming that the pet with a history record has been exposed to a certain allergen multiple times and has experienced allergic reactions and no allergic reactions, then ,in, is the number of exposures, The number of allergic reactions.

3. The cloud architecture clustering management method for pet allergen detection result big data according to claim 2, characterized in that: Step S3 specifically includes: referring to the generalized function set In the form of, construct a neutral virtual pet function set, with a neutral virtual pet function set .

Images