83 results about "Multiple factor analysis" patented technology

The invention relates to the technical field of data processing, and discloses a financial transaction anomaly detection and risk assessment method and device based on artificial intelligence. The method comprises the following steps: collecting and preprocessing multi-source financial transaction related data to obtain a multi-dimensional transaction data set; performing feature engineering and abnormal mode recognition on the multi-dimensional transaction data set to obtain an abnormal transaction recognition result; based on the abnormal transaction identification result and the multi-dimensional transaction data, constructing a financial domain knowledge graph to obtain a relation network between transaction entities; performing multi-dimensional risk assessment on the relation network among the transaction entities to obtain a risk score and risk description data; based on the risk score, performing time sequence analysis and trend extrapolation on the risk description data to obtain a dynamic risk early warning signal; and performing scene simulation and multi-factor analysis on the dynamic risk early warning signal to obtain a risk management strategy suggestion. The efficiency and accuracy of financial transaction anomaly detection and risk assessment based on artificial intelligence are improved.

The invention discloses a slope disease identification method and device based on an unmanned aerial vehicle and a machine vision technology, and the method comprises the steps: carrying out the multi-angle image collection of a slope region on a preset flight path through an unmanned aerial vehicle carrying a high-resolution camera; preprocessing the image, and performing feature extraction and image enhancement; recognizing and classifying disease features by combining an improved YOLOv8-seg instance segmentation algorithm, wherein slope diseases such as cracks, landslides, collapse and the like are covered; quantifying the risk level of the slope disease by using a rule model or a multi-factor analysis method based on the identification result; and carrying out dynamic change prediction on the acquired disease time sequence image by adopting a long-short term memory network (LSTM), and predicting the development trend and the potential instability time of the disease. On the basis of image data collected by the unmanned aerial vehicle, real-time monitoring and dynamic analysis are carried out on the slope diseases, potential hidden danger areas are found in time, disease features can be accurately recognized, and subjective errors possibly caused by manual judgment in a traditional method are avoided.

The invention discloses an intelligent fire-fighting risk prediction and dynamic early warning method and system based on multi-source data fusion, and relates to the technical field of intelligent fire-fighting risk prediction, and the method comprises the steps: collecting multi-source fire-fighting data, and constructing a fire-fighting safety state multi-dimensional data matrix; constructing a risk prediction model based on the fire safety state multi-dimensional data matrix to carry out real-time risk scoring; and executing dynamic early warning through a risk prediction result. According to the method, through the multi-source data acquisition and fusion module, multi-type fire-fighting data are comprehensively acquired and standardized, a structured and multi-dimensional fire-fighting safety state matrix is constructed, and the problems of data dispersion and information splitting are solved. In combination with a risk collaborative prediction and scoring module, the risk level of each region is dynamically output based on multi-factor analysis, and the accuracy and foresight of risk prediction are improved. Through a dynamic early warning generation and linkage module, real-time grading early warning and linkage of a fire fighting system are realized, and rapid response and intelligent prevention and control of a high-risk area are realized.

The invention discloses an operation and maintenance management system and method for a smart city, and belongs to the technical field of electric digital data processing, and the method specifically comprises the steps: automatically collecting and preprocessing city comprehensive management data through employing an intelligent data capturing technology, building a multi-target garbage station planning model according to the set target and constraint conditions, and carrying out the construction of a multi-target garbage station planning model; solving and generating a preliminary planning scheme through an optimization algorithm; in the scheme evaluation stage, a multi-factor analysis method is adopted to comprehensively view the preliminary scheme, and optimization adjustment is carried out according to the preliminary scheme to obtain a final operation and maintenance scheme; a scheme implementation effect is simulated by means of a three-dimensional modeling and simulation technology, and a feasible implementation scheme is formulated in combination with actual conditions; the intelligent planning of the garbage station is realized, and the urban garbage treatment efficiency and the management level are improved.

The invention discloses a treatment response prediction method and system, a storage medium and a terminal, and the method comprises the steps: carrying out the heterogeneity analysis and cell division of first single cell sequencing data of a patient through employing a clustering algorithm, and obtaining a plurality of corresponding cell subgroups; performing feature extraction and biomarker recognition on all the cell subgroups to obtain corresponding cell subgroup features; wherein the cell subset characteristics comprise a cell development trajectory and a plurality of cell markers; and inputting the clinical information and the cell subset characteristics of the patient into the treatment response prediction model, so that the model performs multi-factor analysis and outputs a treatment response prediction result of the patient for each treatment scheme. By adopting the embodiment of the invention, the tiny key difference between the cells is captured, the cells are divided based on the difference between the cells, and then feature extraction and biomarker recognition are performed on each cell subset, so that the specific gene expression modes of different cell subpopulations corresponding to diseases can be recognized, and the treatment response prediction accuracy is further improved.

The invention belongs to the technical field of eye detection, and particularly relates to a myopia progress risk prediction system, which comprises a data acquisition module used for collecting a plurality of biological index data related to myopia progress, including but not limited to ocular axis length, corneal curvature, diopter, age, gender, lifestyle, genetic information and the like; the multi-factor analysis module is used for comprehensively analyzing multiple pieces of biological index data and modeling the individualized myopia progress risk based on a multi-factor interaction relationship; the regression algorithm module is used for collecting multiple pieces of biological index data; by analyzing multiple biological indexes at the same time, comprehensive evaluation of the myopia progress risk is achieved, compared with a traditional method which only depends on a single index for prediction, the system considers interaction of multiple factors, individual differences and environmental influences can be captured more comprehensively, and therefore prediction accuracy and reliability are remarkably improved.

The invention provides a sound insulation performance optimization method of a multilayer film type acoustic metamaterial based on an orthogonal experiment. And a new thought is provided for realizing the design of the low-frequency broadband sound insulation material. Firstly, an MAM unit cell model based on a simplified spider topological structure is designed in combination with an existing polymorphic anti-resonance cooperative type MAM, a three-layer structure is established based on the model, and improvement of the sound insulation index and widening of the sound insulation bandwidth are achieved. Secondly, analyzing the influence of the number of resonators on the sound insulation performance in combination with sound transmission loss (STL), and revealing the sound insulation mechanism through intrinsic modes of different numbers of resonators in combination with the STL. Thirdly, analyzing the sound insulation performance of the structure when the number of the resonators in each layer is different, and screening out the structure with the optimal sound insulation performance; and finally, performing multi-factor analysis on the sound insulation performance by using an orthogonal experiment to obtain an optimal combination of a plurality of parameters, verifying the accuracy, and further optimizing the sound insulation performance.

The invention relates to the field of electrolytic copper production, in particular to a copper electrolysis production quality closed-loop management and control method, which comprises the following steps: collecting cathode copper surface quality data and process parameters in real time, firstly carrying out data cleaning and normalization, and realizing automatic defect classification in combination with univariate probability density analysis; then, adopting Graphical Lasso multivariable association mining to lock core process parameters, and fusing a Bayesian network and a CART decision tree to construct an anomaly prediction model; and dynamically tracing an abnormal root based on a graph computing technology and generating a key path, and finally outputting a data-driven process parameter adjustment suggestion. By applying a single-factor and multi-factor analysis method and combining a decision tree model and a graph calculation technology, abnormal conditions are identified and early warned, factors with the strongest association are found out, reason tracing is carried out, and auxiliary guidance suggestions of process parameters are provided for teams and groups, so that the overall quality of cathode copper in the electrolysis production process is improved.

The present invention provides a method for predicting the risk degree of casing deformation induced by fracturing based on multi-factor analysis, comprising the following steps: determining the influencing factors of casing deformation induced by fracturing; analyzing logging data to identify the above-mentioned influencing factors; defining the risk factors of casing deformation induced by fracturing according to the logging data analysis results; judging the risk degree level of casing deformation induced by fracturing by the number of risk factors existing in the casing, wherein when the number of risk factors is 0, the risk degree is none, when the number of risk factors is 1, the risk degree is the first level, when the number of risk factors is 2, the risk degree is the second level, when the number of risk factors is 3, according to whether the thickness of the casing is greater than 11mm, the risk degree is between the second level and the third level, and when the number of risk factors is 4, the risk degree is the third level. The present invention can extract and identify the risk degree of casing deformation before fracturing and during fracturing, so as to timely optimize the fracturing construction parameters and reduce the probability of casing deformation.

The invention belongs to the technical field of biological information, and provides an ampulla cancer patient prognosis model based on machine learning and a construction method thereof. According to the method, clinical information of an ampulla cancer patient is obtained through a cancer statistical database, clinical characteristics independently related to survival of the patient are analyzed and positioned through single-factor and multi-factor COX, a prognosis model of the ampulla cancer is established through multiple machine learning algorithms according to the clinical characteristics, and the method can be used for survival prediction of the patient. The method has the advantages of being rapid in prediction, high in accuracy and easy and convenient to use, gene determination does not need to be specially conducted on tissue or blood sampling of the patient, accurate survival rate judgment can be achieved only through clinical information, a clinician can make a personalized treatment scheme of the patient according to the prognosis state, and therefore the survival rate of the patient is increased.

The invention discloses a propping agent injection method based on particle size distribution optimization, which comprises the following steps: comprehensively analyzing geometrical morphology parameters of cracks, property parameters of rocks and pressure parameters changing in real time in a fracturing process; determining propping agent particle size distribution ranges corresponding to different areas of the crack through a data analysis model and an algorithm; dividing the particle size distribution range of the proppant particles into a plurality of particle size adaptation intervals according to the width measurement values and permeability test results of different areas of the crack; in the proppant injection process, real-time data monitoring is conducted through an underground sensor, the area with abnormal proppant injection and distribution in the crack is analyzed and judged according to real-time data, and dynamic particle size adjustment is conducted. According to the method, the problem that a traditional propping agent injection mode is uneven in distribution in a complex underground environment is solved, key factors such as the geometrical morphology of a crack, rock properties and fracturing pressure are accurately controlled by constructing a multi-factor analysis model, and fine regulation and control of particle size distribution of the propping agent are achieved.

The invention relates to a method and system for evaluating acceleration factors influencing lithium ion battery aging, and the method comprises the steps: carrying out the parameter setting of a plurality of acceleration factors, and obtaining the test conditions of different acceleration factor combinations; under the test condition, performing an aging test on the lithium ion battery and performing real-time monitoring to obtain a battery aging process data set; and performing multi-factor analysis on the battery aging process data set to obtain influence degree evaluation results of different acceleration factors on the battery aging rate. According to the method, multiple acceleration factors are considered, and the comprehensiveness and accuracy of the battery aging test are remarkably improved.

Backup data is leveraged to determine whether primary data has been encrypted by malware. The disclosed approach does not rely on recognizing particular malware instances or malware provenance, and thus can be applied to any body of data. Even a novel and previously unknown malware attack can be detected in this way. An illustrative data storage management system analyzes secondary copies it created over time, applies a multi-factor analysis to data recovered from the secondary copies and, based on the analysis, infers whether the primary data from which the secondary copies were created may be encrypted. The present approach uses successive versions of backup copies to find indicia of malware encryption, rather than trying to trace or identify the malware itself. Indicia of entropy correlate highly with encryption, such as encryption performed by malware attacks. Conversely, indicia of similarity correlate highly with lack of encryption of successive versions of documents.

The invention discloses a power demand prediction method based on time sequence and external environment factors, relates to the technical field of data processing and prediction, and solves the problem that power potential cannot be systematically evaluated due to the fact that no system performs data analysis and prediction on a power market in the prior art. Customer basic information, business expansion information, power consumption behaviors and external environment factors are obtained, then a user business expansion influence power consumption demand multi-factor analysis model is constructed according to business expansion business subdivision, and a user business expansion influence power consumption demand multi-factor analysis model is established by combining the industry prosperity, the electric quantity change rate in the power behaviors, the capacity change amount in the customer basic information and other multi-dimensional information. According to the method, a power demand coefficient of a comprehensive evaluation property is obtained, then an ARIMA model is adopted for fitting prediction of the power demand coefficient of each business expansion business type in the future, prediction of the electric quantity potential of cities, industries and enterprises is realized, and a more accurate decision basis is provided for power industry management and enterprise operation.

This invention relates to the field of data processing technology, specifically to a method for evaluating the effectiveness of fish passage in an eco-friendly fishway based on multi-factor analysis. The method involves acquiring multi-dimensional data, a multi-factor evaluation calculation model, and engineering parameters of each batch of fish swimming through a test section within each monitoring period. In each period, the theoretical environmental pressure rate is calculated using the model based on water temperature and flow velocity sequences. The total time consumed is compared with a preset baseline time to determine the energy consumption index. Effective swimming weights are determined by combining the deviation of the swimming trajectory time series and engineering parameters, reducing the weight of ineffective time consumption data in the backwater rest area. Multiple indicators from all batches of fish are integrated to determine long-term evaluation deviation indicators, and effective error periods are selected. An environmental parameter constraint space is constructed, and relevant data from each batch of fish within the effective error period are used for traversal optimization to eliminate data contamination in the rest area and improve the model's adaptability and predictive accuracy to different environments and fish states.

The embodiment of the invention provides a lease fee determination method and system and a storage medium, and belongs to the technical field of budget management. The method comprises the following steps: acquiring depreciation cost and interest cost of stock assets; determining depreciation cost and interest cost of newly added assets based on a multi-factor modeling mode according to historical data of the lease contract and external market environment data; and determining the lease expense in an attribution analysis mode according to a set adjustment factor, the depreciation expense and the interest expense of the newly added assets and the depreciation expense and the interest expense of the stock assets. According to the technical scheme, related cost prediction is adjusted through attribution analysis and according to multi-factor analysis, and the method has the advantage that budget cost planning of the next year can be more accurately carried out.

The invention discloses an application of Hsa-miR-24-3p in construction of a pancreatic ductal adenocarcinoma postoperative total lifetime prediction model. According to the invention, the expression of Hsa-miR-24-3p in preoperative serum of a PDAC patient is analyzed, and the total lifetime of a patient with increased level of Hsa-miR-24-3p (miR-24) is obviously reduced (Plt; 0.0001) of the method for manufacturing the same; multi-factor analysis shows that miR-24 is a prognosis factor independent of the age, gender and clinical pathological characteristics of a patient, and the prediction efficiency of miR-24 is remarkable; the area under the ROC curve reaches 0.82, which is superior to the area under the CA19-9 (0.61), the area under the CA125 (0.59), the area under the CA50 (0.51) and the area under the CEA (0.56); when the miR-24 and the TNM are staging, and the CA19-9 and the CA125 are integrated into the column diagram model, the prognosis prediction accuracy is further improved.

The invention provides a construction method and equipment of a fracture risk prediction model, a medium and a program product, and relates to the field of intelligent medical treatment. The method comprises the following steps: acquiring basic data of a T2DM osteoporosis patient as a training set sample; the training set samples are divided into a fracture group and a fracture-free group; performing statistical analysis on the basic data, representing continuous variables conforming to normal distribution by using mean + / -standard deviation, and comparing the difference between the two groups through independent sample t inspection; adopting median and quartile to describe continuous variables which do not conform to normal distribution; through single-factor and multi-factor analysis, screening fracture influence characteristics of the osteoporosis patient from the basic data; and constructing a prediction model of the fracture risk of the osteoporosis patient based on the prediction variables. A special group of postmenopausal female diabetes is used as a research object, difference characteristics are analyzed, a prediction model is constructed, and early prevention and treatment of osteoporosis fracture patients are achieved.

The invention provides a young triple negative breast cancer prognosis prediction method based on machine learning, belongs to the technical field of medical diagnosis, and aims to improve accurate evaluation of prognosis of young triple negative breast cancer patients through an intelligent prediction model. According to the method, a data set is constructed based on a SEER database, and the data set comprises demographic characteristics, tumor grading, TNM staging and treatment related information (such as surgery, radiotherapy, chemotherapy and the like) of a patient. The data integrity and consistency are ensured by preprocessing a data set, including missing value filling, unbalanced data processing and other operations; and screening variables based on single-factor and multi-factor analysis, and evaluating the correlation between each feature and prognosis. Then, model construction is carried out based on various classification models (such as XGBoost, GBM and random forest), and an optimal prediction model (GBM) is determined through hyper-parameter optimization and cross validation; in addition, the optimal model is globally and locally explained by adopting interpretable methods such as Permutation Implement and LIME, and key prognosis characteristics and contribution of the key prognosis characteristics to a prediction result are determined. By means of the auxiliary decision making system, a clinician can visually understand a model prediction result, and the scientificity of diagnosis and treatment scheme making is improved. According to the invention, a more reliable personalized prognosis evaluation tool can be provided for young triple negative breast cancer patients, and the long-term survival prognosis of the patients can be improved.

The invention discloses a garment production order accurate quotation system based on multi-factor analysis, which belongs to the technical field of garment production and comprises a data acquisition module, a material unit price setting module, a rule coefficient setting module, a quotation calculation module and a quotation output module. According to the invention, the real-time unit price of the material is formulated based on the collected market dynamic data, then the real-time coefficient of the customer and the real-time coefficient of the order are formulated based on the collected customer related information data and the order feature information data, and then a plurality of order quotations are calculated based on the formulated real-time unit price of the material, the real-time coefficient of the customer and the real-time coefficient of the order. Compared with the prior art, the order quotation can be quickly obtained, adjustment can be performed according to the real-time condition of the market, and the rationality and timeliness of quotation are ensured.

The present invention discloses a proppant injection method based on particle size distribution optimization, comprising: comprehensively analyzing the geometric parameters of the fracture, the property parameters of the rock, and the pressure parameters that change in real time during the fracturing process, and determining the proppant particle size distribution range corresponding to different areas of the fracture through a data analysis model and algorithm; dividing the proppant particle size distribution range into a plurality of particle size adaptation intervals according to the width measurement values ​​and permeability test results of different areas of the fracture; during the proppant injection process, using downhole sensors to perform real-time data monitoring, and judging the areas in the fracture where proppant injection and distribution abnormalities exist based on the real-time data analysis, and performing dynamic particle size adjustment. The present invention solves the problem of uneven distribution of traditional proppant injection methods in complex underground environments. By constructing a multi-factor analysis model, key factors such as fracture geometry, rock properties, and fracturing pressure are accurately controlled to achieve refined regulation of proppant particle size distribution.

The invention provides a wind power plant fan optimization arrangement system and method based on multi-factor analysis, and relates to the technical field of data optimization, and the system comprises a data collection module which collects wind speed data and topographic slope data of each monitoring point in a wind power plant region, and builds a basic data set; the region determination module is used for extracting core performance parameters of a target fan, performing threshold screening on wind speed data and slope values in the basic data set, determining an effective wind energy region and a gentle terrain region, and further generating a candidate arrangement region; the area analysis module is used for determining the minimum spacing required by operation of a single fan according to the blade size, and determining the maximum number upper limit of the fans capable of being arranged in each area in combination with the area and the shape of the candidate arrangement area; the scheme determination module is used for generating a plurality of groups of arrangement schemes for each candidate arrangement area; and the scheme screening module is used for determining a final fan optimization arrangement scheme based on all the schemes. The overall power generation efficiency of the wind power plant is improved, and the energy loss is reduced.

The application discloses a kind of geophysical response characterization methods and systems of element enrichment degree in coal rock, and characterization method includes the following steps: obtaining the original sample of the same research block coal rock, and dividing subdivision layer;The sample of the subdivision layer is processed to obtain a variety of sizes of cube samples and its remaining material to meet different test conditions;Collect remaining material and acidolysis into solution, select target element and determine the element enrichment degree of acidolysis solution;Determine the multiple geophysical responses of the cube sample;Calculate geophysical derived parameters;According to the test and calculation of the above steps, a plurality of geophysical responses corresponding to the element content are obtained for a certain number of subdivision layer samples, and a geophysical response characterization model of element enrichment degree is constructed using a multi-factor analysis method;Different characterization models constructed by different multi-factor analysis methods are optimized, and the model with high precision and easy operation is determined as the final characterization model.

The invention relates to a bromhidrosis postoperative wound infection risk prediction method and a bromhidrosis postoperative wound infection risk prediction system, aims to solve the problems that an existing clinical evaluation method depends on subjective experience and lacks a standardized tool, and performs quantitative prediction on the individual infection risk of a patient by utilizing a column graph model constructed based on multi-factor analysis. The method comprises the following steps: systematically collecting clinical index data of a patient, wherein the clinical index data comprises general data and operation-related indexes; inputting the data into a pre-constructed column graph model, and obtaining a comprehensive risk score by matching a corresponding score for each index and performing accumulation; and finally, the risk level is divided into a low-risk class and a high-risk class according to a preset threshold value, and an intuitive basis is provided for clinical decision making. According to the method, the operation process is standardized, the result output is visual, the method can be effectively integrated into the clinical working process, and a practical auxiliary tool is provided for medical staff to identify high-risk patients and implement targeted prevention and nursing measures, so that the quality and efficiency of postoperative management are expected to be improved.

The invention discloses a construction project planning site selection and land use pre-examination evaluation method based on multi-factor analysis, and particularly relates to the technical field of construction project planning, and the method comprises the following steps: firstly, collecting basic factor, environmental factor and social factor data related to a project; then independently analyzing the basic factor data, calculating deviation values and combining the deviation values to form a deviation vector; carrying out predictive quantitative evaluation on the environmental factor data, and calculating an environmental influence index; performing qualitative analysis on the social factor data to generate a social influence index; and finally, inputting the basic factor deviation vector, the environmental influence index and the social influence index into a pre-trained multi-factor decision model, classifying layer by layer, and outputting an evaluation result to provide a scientific basis for site selection. According to the method, the limitation of a traditional single-factor decision-making method is overcome, the suitability of the site selection area is comprehensively evaluated, the result is more persuasive due to the interpretability and transparency of the decision-making process, and project approval and communication are facilitated.

The invention provides a tool for intelligently evaluating the dystrophy risk of a patient with nasopharyngeal carcinoma, and the tool for evaluating the specific dystrophy of nasopharyngeal carcinoma is constructed by integrating indexes such as social demographic data, physique measurement, tumor staging, chemoradiotherapy conditions, recent adverse reactions and hematology indexes, so that high-precision risk evaluation is realized. Through single-factor and multi-factor analysis, the variable weight is optimized, and the discrimination capability and stability of the model are improved. An intelligent applet based on artificial intelligence is developed, patient data are extracted, nutrition risk calculation is completed within seconds, evaluation time is shortened, and clinical working efficiency is improved. The system also has the functions of dynamic monitoring and intelligent early warning, automatically updates the evaluation result along with the treatment progress of the patient, and provides individualized intervention suggestions. The defects that a traditional nutrition evaluation tool is high in subjectivity, low in efficiency, lack of tumor specificity and low in informatization degree are overcome, and an efficient, accurate and automatic intelligent evaluation solution is provided for clinic.

The invention is suitable for the technical field of data analysis, and particularly relates to an urban spatial form data analysis method and system based on machine learning, and the method comprises the steps: determining urban to-be-analyzed data of an urban spatial form; wherein the urban to-be-analyzed data is used for indicating a traffic infrastructure area in an urban spatial form; performing multi-factor analysis on the to-be-analyzed data to obtain factor influence data; wherein the factor influence data is used for reflecting the traffic influence condition of multiple factors on the traffic infrastructure area; generating factor result data of the urban spatial form according to the factor influence data; wherein the factor result data is used for indicating a solution obtained according to the traffic influence condition. According to the urban spatial form data analysis method based on machine learning provided by the invention, the traffic demands under different conditions can be met, the traffic jam condition is reduced, and the traffic efficiency is improved.

A software and / or hardware facility for scoring the quality of traffic to a site accessible via the Internet or other network. The facility extracts data from one or more data logs that are obtained from a publisher, advertiser, or third party. A multi-factor analysis in the form of a rule set is applied by the facility to the data. The analysis of the data identifies agent actions that are desirable to a publisher, advertiser, or third party. The facility generates a relative score for each agent action. The score may be used to assess the quality of traffic received by a network site. Lower scores are indicative of traffic having little value, whereas higher scores are indicative of traffic having greater value. The score may be provided to an advertising network and used to charge a variable amount for advertisements based on the quality of traffic that the advertisements receive.

The application discloses a water body peculiar smell olfactory traceability method and device based on multi-factor analysis and a medium, and can be applied to the water body peculiar smell traceability technical field. After the peculiar smell olfactory module, the olfactory unit and the odor sampling unit which interact with the control end are respectively arranged at appropriate positions of each monitoring water area, the olfactory unit is controlled to perform odor olfaction on the monitoring water area, when the odor olfaction result shows that the odor of the corresponding monitoring water area is abnormal, the odor sampling unit in the monitoring water area is controlled to perform gas sampling to obtain a gas sample, then the target substance information detection result of the peculiar smell corresponding to the odor sample is obtained, the water quality sampling module in all the monitoring water areas is controlled to perform water body sampling to obtain a water body sample, then the water sample detection data of the odor-causing target substance is obtained, and finally, the water body peculiar smell olfactory traceability is performed according to the water sample detection data, so that the source position of the water body peculiar smell can be accurately found.

The invention provides a multi-modal intelligent security event analysis method and system, and the method comprises the steps: obtaining monitoring data in a monitoring region, the monitoring data comprising sensing data collected by a plurality of sensors and video data collected by a plurality of cameras; when an alarm signal sent by a target sensor is received, calculating a weighted score of the alarm event through a composite weighting algorithm based on the equipment attribute of the target sensor, historical alarm data, whether a collaborative alarm signal exists or not and a current environmental factor; carrying out feature extraction and classification on the video data of the region where the target sensor is located by adopting a two-section multi-factor analysis technology so as to identify whether an abnormal behavior exists or not; and determining the risk level of the alarm event according to the weighted score and the identified abnormal behavior, and executing a corresponding grading response operation according to the risk level, thereby reducing emergency resource waste caused by false alarm.