Enterprise financial risk early warning system and method based on big data

Abstract

The invention provides an enterprise financial risk early warning system and method based on big data, and belongs to the technical field of data analysis and processing, the system comprises a data acquisition module, a data processing module, a risk assessment module, an early warning module, a logistic regression algorithm module and the like, and the data acquisition module comprehensively collects internal and external data of an enterprise; the data processing module performs data cleaning, integration and standardization; the risk assessment module selects proper indexes and constructs various models; the scheme aims to solve the problems of incomplete data acquisition, insufficient data processing capability, imperfect risk assessment model, insensitive early warning mechanism and the like in the prior art, and can realize comprehensive data acquisition, efficient data processing, accurate risk assessment and sensitive early warning mechanism. The financial risk coping capability of enterprises is improved, and stable development of the enterprises is guaranteed.

Description

Technical Field

[0001] The present invention belongs to the technical field of data analysis and processing, and specifically relates to an enterprise financial risk early warning system and method based on big data. Background Art

[0002] In today's business operations, financial risk is one of the important challenges faced by enterprises. With the development of big data technology, enterprises can obtain and process a large amount of financial data. However, in traditional financial management, it is often difficult to effectively use this data to conduct accurate financial risk warnings. At present, many enterprises have the following problems in financial risk warnings:

[0003] Incomplete data collection: focusing only on the company's internal financial data, while ignoring important information such as external market dynamics, industry trends, and competitors' financial performance, resulting in an incomplete assessment of risks;

[0004] Insufficient data processing capabilities: Unable to effectively clean, integrate and standardize massive amounts of data, resulting in low data quality, which affects the accuracy of subsequent risk assessments and early warnings;

[0005] Imperfect risk assessment model: The risk assessment methods and models used are relatively simple and fail to fully consider the combined impact of multiple financial indicators and non-financial factors, making it difficult to accurately assess the financial risk level of the enterprise;

[0006] The early warning mechanism is not sensitive: the generation and transmission of early warning signals are not timely enough, and the early warning method is single, which cannot meet the needs of enterprises to respond to risks quickly, resulting in delayed response of enterprises when facing risks, which may cause greater losses;

[0007] This solution aims to solve the above problems and provide a corporate financial risk early warning system and method based on big data; by comprehensively collecting internal and external data of the enterprise, using advanced data processing technology and diversified risk assessment models, it can achieve accurate assessment and timely early warning of corporate financial risks;

[0008] Specifically, the system of this solution includes data acquisition module, data processing module, risk assessment module, early warning module and logistic regression algorithm, etc., which can effectively integrate and analyze data, improve the accuracy of risk assessment and the timeliness of early warning, help enterprise management to better formulate response strategies, reduce financial risks and ensure the stable development of the enterprise. Summary of the invention

[0009] The purpose of the present invention is to provide an enterprise financial risk early warning system and method based on big data, aiming to solve the problems in the prior art such as incomplete data collection, neglect of important external information, incomplete risk judgment, insufficient data processing capabilities, low data quality, affecting the accuracy of risk assessment and early warning, imperfect risk assessment model, difficulty in accurately assessing the level of enterprise financial risk, insensitive early warning mechanism, untimely generation and transmission of early warning signals, and single method, which leads to delayed response of enterprises to risks.

[0010] To achieve the above object, the present invention provides the following technical solutions:

[0011] The enterprise financial risk early warning system and method based on big data is characterized by including:

[0012] An enterprise financial risk early warning system based on big data, the enterprise financial risk early warning system based on big data comprising a data acquisition module, a data processing module, a risk assessment module, an early warning module and a logistic regression algorithm;

[0013] The data collection module includes an internal data collection unit, a financial statement collection unit, an account detail collection unit, a budget planning collection unit, a fund flow record collection unit, an external data collection unit, an industry dynamics collection unit, a market trend collection unit, and a competitor financial performance collection unit;

[0014] Data processing module: the data cleaning unit, duplicate data removal unit, error data correction unit, incomplete data supplement unit, data integration unit, multi-data source fusion unit, data standardization unit, format unification unit, unit conversion unit;

[0015] Risk assessment module: the indicator selection unit, financial ratio indicator selection unit, non-financial indicator selection unit, model building unit, multivariate linear regression model building unit, logistic regression model building unit, decision tree model building unit, neural network model building unit, model verification and optimization unit,

[0016] Early warning module: The early warning module includes an early warning indicator unit, a financial ratio indicator threshold setting unit, a market indicator threshold setting unit, a non-financial indicator threshold setting unit, an early warning method selection unit, a real-time pop-up reminder unit, an email notification unit, an SMS alarm unit, an early warning report generation unit, a risk type presentation unit, and a risk level assessment unit.

[0017] As a preferred solution of the present invention, step 1, data collection: collecting the internal financial data of the enterprise, including financial statements, account details, budget planning and fund flow records, etc., and capturing external data, such as industry dynamics, market trends, competitors' financial performance and macroeconomic data, etc.;

[0018] Step 2: Data preprocessing: clean the collected data, remove duplicate, erroneous and incomplete data, integrate data from different sources to form a unified data format, standardize the data, and unify the format and unit of the data for subsequent analysis;

[0019] Step 3: Determine risk indicators: Select appropriate financial ratio indicators, such as debt-to-asset ratio, current ratio, net profit growth rate, etc., and include non-financial indicators, such as market share changes, customer satisfaction, supplier stability, etc.;

[0020] Step 4: Risk assessment model construction: Build a model based on multiple linear regression, logistic regression, decision tree or neural network methods, and use historical data to train and optimize the model;

[0021] Step 5: Risk assessment and prediction: Input the preprocessed data into the model to assess the company’s current financial risk level and predict future financial risks;

[0022] Step 6: Set warning thresholds: Determine the thresholds for different risk indicators based on the enterprise’s risk tolerance and industry characteristics;

[0023] Step 7: Early warning signal generation: When the risk indicator exceeds the threshold, a corresponding early warning signal is generated;

[0024] Step 8: Early warning transmission and response: Early warning signals are transmitted to relevant personnel through various means, such as real-time pop-up windows, emails, and text messages. The company management formulates response strategies based on the early warning signals and takes measures to reduce risks.

[0025] Step 9: Model monitoring and updating: Continuously monitor the accuracy and effectiveness of the model, and update and optimize the model based on new data and business changes.

[0026] Compared with the prior art, the present invention has the following beneficial effects:

[0027] 1. In this solution, first, data collection is comprehensive: through the data collection module, not only the internal financial statements, account details, budget planning and capital flow records of the enterprise are collected, but also the external industry dynamics, market trends, competitor financial performance and macroeconomic data are captured, which provides a rich source of information for the comprehensive assessment of enterprise risks and makes risk judgment more comprehensive;

[0028] 2. Efficient data processing: Units such as data cleaning, deduplication, correction of erroneous data and supplementation of incomplete data in the data processing module can improve the quality and accuracy of data; data integration and multi-data source fusion units integrate data from different sources into a unified data set for subsequent analysis; data standardization units ensure the comparability and consistency of data, all of which help to improve the efficiency and quality of data processing, thereby providing a reliable data basis for accurate risk assessment and early warning.

[0029] 2. In this solution, 1. Precision of risk assessment: The risk assessment module uses multiple linear regression, logistic regression, decision tree, neural network and other methods to build a model, and selects appropriate financial ratio indicators and non-financial indicators, which can fully consider the combined impact of multiple factors, so as to more accurately assess the financial risk level of the enterprise.

[0030] 2. Sensitization of early warning mechanism: The early warning module sets a variety of early warning indicators and thresholds, and timely transmits early warning signals through various means such as real-time pop-up reminders, email notifications and SMS alerts, so that the company's management can formulate response strategies in time and take measures to reduce risks, effectively avoiding losses caused by untimely early warnings. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention.

[0032] In the attached picture:

[0033] Figure 1 This is a flow chart of the enterprise financial risk early warning system and method based on big data of the present invention;

[0034] Figure 2 The first process distribution diagram of the logistic regression algorithm of the present invention is

[0035] Figure 3 This is a second process distribution block diagram of the logistic regression algorithm of the present invention. DETAILED DESCRIPTION

[0036] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0037] Example 1

[0038] See also Figure 1-Figure 3, the present invention provides the following technical solutions:

[0039] The enterprise financial risk early warning system and method based on big data is characterized by including:

[0040] An enterprise financial risk early warning system based on big data, the enterprise financial risk early warning system based on big data comprising a data acquisition module, a data processing module, a risk assessment module, an early warning module and a logistic regression algorithm;

[0041] The data collection module includes an internal data collection unit, a financial statement collection unit, an account detail collection unit, a budget planning collection unit, a fund flow record collection unit, an external data collection unit, an industry dynamics collection unit, a market trend collection unit, and a competitor financial performance collection unit;

[0042] Data processing module: the data cleaning unit, duplicate data removal unit, error data correction unit, incomplete data supplement unit, data integration unit, multi-data source fusion unit, data standardization unit, format unification unit, unit conversion unit;

[0043] Risk assessment module: the indicator selection unit, financial ratio indicator selection unit, non-financial indicator selection unit, model building unit, multivariate linear regression model building unit, logistic regression model building unit, decision tree model building unit, neural network model building unit, model verification and optimization unit,

[0044] Early warning module: The early warning module includes an early warning indicator unit, a financial ratio indicator threshold setting unit, a market indicator threshold setting unit, a non-financial indicator threshold setting unit, an early warning method selection unit, a real-time pop-up reminder unit, an email notification unit, an SMS alarm unit, an early warning report generation unit, a risk type presentation unit, and a risk level assessment unit.

[0045] In a specific embodiment of the present invention,

[0046] Enterprise financial risk early warning system based on big data:

[0047] The system is the core of the entire early warning system, integrating multiple modules such as data collection, processing, risk assessment and early warning, as well as logistic regression algorithms, aiming to comprehensively and accurately warn of corporate financial risks;

[0048] Data acquisition module:

[0049] Internal data collection unit: responsible for collecting financial data within the enterprise, such as financial statements, account details, budget planning and capital flow records, etc. These data reflect the financial status and operation of the enterprise itself;

[0050] Financial Statement Collection Unit: Specialized in collecting the financial statements of enterprises, including balance sheets, income statements and cash flow statements, etc. These statements are important bases for evaluating the financial status of enterprises;

[0051] Account details collection unit: Obtaining the specific account details of the enterprise helps to gain a deeper understanding of the enterprise's financial transactions and capital flows;

[0052] Budget planning collection unit: collects the company's budget planning information, helps compare the differences between actual operations and budgets, and promptly identifies potential risks;

[0053] Fund flow record collection unit: pays attention to the inflow and outflow of funds of the enterprise and monitors the liquidity risk of funds;

[0054] External data collection unit: obtain relevant data from outside the enterprise, such as industry dynamics, market trends and competitor financial performance, to provide an external perspective for comprehensive assessment of enterprise risks;

[0055] Industry dynamics collection unit: collects the latest industry dynamics, policy changes and other information to help companies understand the development trend of their industry;

[0056] Market trend collection unit: capture market trend information, such as changes in market demand, price fluctuations, etc., so that enterprises can adjust their business strategies in a timely manner;

[0057] Competitor financial performance collection unit: focuses on competitors' financial status and business strategies, providing a basis for competitive analysis for enterprises;

[0058] Data processing module:

[0059] Data cleaning unit: removes duplication, errors and incomplete parts in the data to improve the quality and accuracy of the data;

[0060] Remove duplicate data units: Specialize in processing duplicate records in data to ensure the uniqueness of data;

[0061] Correction of error data unit: Correction of erroneous data to ensure data reliability;

[0062] Supplement incomplete data units: fill in the missing parts of the data to make the data more complete;

[0063] Data integration unit: Integrate data from different data sources to form a unified data set for subsequent analysis;

[0064] Multi-data source fusion unit: realizes the fusion of multiple data sources and makes full use of various data resources;

[0065] Data standardization unit: standardize data, including format unification and unit conversion, to ensure data comparability and consistency;

[0066] Format unification unit: unify the data format into a standard format to facilitate data storage and processing;

[0067] Unit conversion unit: convert the units in the data so that data in different units can be analyzed uniformly;

[0068] Risk Assessment Module:

[0069] Indicator Selection Unit: Responsible for selecting appropriate financial ratio indicators and non-financial indicators to comprehensively assess the financial risks of the enterprise;

[0070] Financial ratio indicator selection unit: select financial ratio indicators such as debt-to-asset ratio, current ratio, net profit growth rate, etc. These indicators can reflect the financial health of the enterprise;

[0071] Non-financial indicator selection unit: Incorporate non-financial indicators such as market share changes, customer satisfaction, and supplier stability to assess corporate risks from multiple perspectives;

[0072] Model building unit: Use multiple linear regression, logistic regression, decision tree, neural network and other methods to build risk assessment models;

[0073] Multiple linear regression model building unit: Establish multiple linear regression models to analyze the linear relationship between multiple variables and financial risks;

[0074] Logistic regression model building unit: builds a logistic regression model, which is suitable for binary classification problems and can determine whether the company faces financial risks;

[0075] Decision tree model building unit: Use decision tree model to conduct risk assessment, which is intuitive and easy to understand;

[0076] Neural network model building unit: Use neural network models to process complex nonlinear relationships and improve the accuracy of risk assessment;

[0077] Model verification and optimization unit: verify and optimize the constructed model to ensure the accuracy and reliability of the model;

[0078] Early warning module:

[0079] Early warning indicator unit: determine the indicators used for early warning, including financial ratio indicators, market indicators and non-financial indicators;

[0080] Financial ratio indicator threshold setting unit: sets the threshold of the financial ratio indicator and triggers an early warning when the indicator exceeds the threshold;

[0081] Market indicator threshold setting unit: sets corresponding thresholds for market indicators to timely detect market risks;

[0082] Non-financial indicator threshold setting unit: determine the warning threshold of non-financial indicators, taking into account the impact of multiple factors;

[0083] Warning method selection unit: select appropriate warning methods, such as real-time pop-up reminders, email notifications, and SMS alerts;

[0084] Real-time pop-up reminder unit: Send early warning information to relevant personnel through real-time pop-up windows to ensure timely awareness of risks;

[0085] Email notification unit: Use email to notify relevant personnel and provide detailed warning information and analysis reports;

[0086] SMS alarm unit: Sends alarm information via SMS, so that relevant personnel can take timely measures;

[0087] Early warning report generation unit: Generates early warning reports, including risk type presentation, risk level assessment, etc., to provide a basis for decision-making;

[0088] Risk type presentation unit: clearly present the risk types faced by the enterprise, such as market risk, credit risk, etc.;

[0089] Risk Assessment Unit: Assess the degree of risk and help enterprises understand the severity of the risk;

[0090] The specific logistic regression algorithm can effectively process multi-dimensional data, and incorporate the company's financial ratio indicators (such as debt-to-asset ratio, current ratio, net profit growth rate, etc.) and non-financial indicators (such as market share changes, customer satisfaction, supplier stability, etc.) into the analysis, so as to more comprehensively evaluate the financial risks faced by the company. Through training on historical data, the algorithm can learn the complex relationship between different indicators and financial risks, and achieve accurate risk prediction. The logistic regression algorithm can adapt to the data characteristics in the big data environment, can process large-scale data samples, and has a certain robustness to noise and outliers in the data, which makes it perform well in processing massive financial data of the company, and can accurately mine the potential patterns and laws in the data. In this system, the logistic regression algorithm can quickly predict and analyze new data and discover potential financial risks in a timely manner. It has high computational efficiency and can meet the real-time requirements of enterprises for risk warnings, helping enterprises respond in the shortest time and reduce losses caused by risks.

[0091] See also Figure 1-Figure 3,Step 1,Data Collection: Collect the internal financial data of the enterprise, including financial statements, account details, budget planning and capital flow records, etc., and capture external data, such as industry dynamics, market trends, competitors' financial performance and macroeconomic data;

[0092] Step 2: Data preprocessing: clean the collected data, remove duplicate, erroneous and incomplete data, integrate data from different sources to form a unified data format, standardize the data, and unify the format and unit of the data for subsequent analysis;

[0093] Step 3: Determine risk indicators: Select appropriate financial ratio indicators, such as debt-to-asset ratio, current ratio, net profit growth rate, etc., and include non-financial indicators, such as market share changes, customer satisfaction, supplier stability, etc.;

[0094] Step 4: Risk assessment model construction: Build a model based on multiple linear regression, logistic regression, decision tree or neural network methods, and use historical data to train and optimize the model;

[0095] Step 5: Risk assessment and prediction: Input the preprocessed data into the model to assess the company’s current financial risk level and predict future financial risks;

[0096] Step 6: Set warning thresholds: Determine the thresholds for different risk indicators based on the enterprise’s risk tolerance and industry characteristics;

[0097] Step 7: Early warning signal generation: When the risk indicator exceeds the threshold, a corresponding early warning signal is generated;

[0098] Step 8: Early warning transmission and response: Early warning signals are transmitted to relevant personnel through various means, such as real-time pop-up windows, emails, and text messages. The company management formulates response strategies based on the early warning signals and takes measures to reduce risks.

[0099] Step 9: Model monitoring and updating: Continuously monitor the accuracy and effectiveness of the model, and update and optimize the model based on new data and business changes.

[0100] In this embodiment: Step 1, data collection: The internal financial data of the enterprise is the basis for assessing financial risks. Financial statements provide important information such as the assets, liabilities, income and expenditure of the enterprise; the account details record the specific circumstances of each financial transaction, which helps to gain an in-depth understanding of the capital flow and financial status of the enterprise; the budget plan reflects the expected income and expenditure of the enterprise. By comparing the actual situation with the budget, potential deviations and risks can be discovered; the capital flow record focuses on the cash inflow and outflow of the enterprise and monitors the liquidity risk of funds. The capture of external data can provide a more comprehensive perspective for the enterprise. Industry dynamics include changes in industry policies and the evolution of the market competition pattern. These factors may have a profound impact on the financial status of the enterprise; market trends such as changes in market demand and price fluctuations are directly related to the sales revenue and profitability of the enterprise; the financial performance of competitors can help enterprises understand their position and competitive advantages in the market; macroeconomic data such as economic growth rate, interest rate, exchange rate, etc. will have a macro-level impact on the business environment of the enterprise;

[0101] Step 2, data preprocessing: Data cleaning is to ensure the quality of data. Removing duplicate data can avoid data redundancy and improve the efficiency and accuracy of analysis. Correcting erroneous data can correct erroneous information in the data and prevent these errors from misleading subsequent analysis. Supplementing incomplete data can make the data more complete and reduce analysis bias caused by missing data. Data integration integrates data from different sources to form a unified data view for comprehensive analysis. A unified data format helps with data storage, management and analysis. Data standardization unifies the format and unit of data to make different indicators comparable, which is convenient for comprehensive evaluation and analysis.

[0102] Step 3, determine the risk indicators: financial ratio indicators such as debt-to-asset ratio reflect the company's debt level and debt repayment ability, the current ratio measures the company's short-term debt repayment ability, and the net profit growth rate reflects the company's profitability and development potential. Non-financial indicators such as market share changes reflect the company's competitive position in the market. Customer satisfaction is related to the company's market reputation and sales stability. Supplier stability affects the company's supply chain stability. Step 4, build a risk assessment model: the multivariate linear regression model can analyze the linear relationship between multiple variables and financial risks, helping companies understand the comprehensive impact of various factors on financial risks. The logistic regression model is suitable for binary classification problems and can accurately determine whether the company is facing financial risks. The decision tree model is intuitive and easy to understand, which can help companies make decisions quickly. The neural network model can handle complex nonlinear relationships and improve the accuracy of risk assessment;

[0103] Step 5: Risk assessment and prediction: Inputting the pre-processed data into the model can assess the company's current financial risk level, identify potential risk points in a timely manner, and predict future financial risks, which will help the company formulate response strategies in advance and reduce the impact of risks;

[0104] Step 6. Set warning thresholds: Determine the thresholds for different risk indicators based on the enterprise's risk tolerance and industry characteristics. Enterprises with strong risk tolerance may set higher thresholds, while enterprises with weaker risk tolerance may set lower thresholds. Industry characteristics will also affect the setting of thresholds. Different industries have different risk levels and characteristics, so they need to be reasonably set according to the actual situation of the industry.

[0105] Step 7: Early warning signal generation: When the risk indicator exceeds the threshold, the system will automatically generate a corresponding early warning signal to promptly remind relevant personnel to pay attention to the risk. The generation of early warning signals needs to be accurate and timely so that the enterprise can quickly take countermeasures;

[0106] Step 8: Early warning transmission and response: Early warning signals are transmitted to relevant personnel through various means, such as real-time pop-up windows, emails, text messages, etc., to ensure that information can be delivered in a timely manner. The company's management formulates response strategies based on early warning signals and takes measures to reduce risks, such as adjusting business strategies, optimizing capital allocation, and strengthening internal controls.

[0107] Step 9: Model monitoring and updating: Continuously monitor the accuracy and effectiveness of the model, promptly identify problems and deficiencies in the model, update and optimize the model based on new data and business changes, ensure that the model can adapt to the development and changes of the enterprise, and provide accurate risk warnings.

[0108] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An enterprise financial risk early warning system and method based on big data, characterized in that: include: An enterprise financial risk early warning system based on big data, the enterprise financial risk early warning system based on big data comprising a data acquisition module, a data processing module, a risk assessment module, an early warning module and a logistic regression algorithm; The data collection module includes an internal data collection unit, a financial statement collection unit, an account detail collection unit, a budget planning collection unit, a fund flow record collection unit, an external data collection unit, an industry dynamics collection unit, a market trend collection unit, and a competitor financial performance collection unit; Data processing module: the data cleaning unit, duplicate data removal unit, error data correction unit, incomplete data supplement unit, data integration unit, multi-data source fusion unit, data standardization unit, format unification unit, unit conversion unit; Risk assessment module: the indicator selection unit, financial ratio indicator selection unit, non-financial indicator selection unit, model building unit, multivariate linear regression model building unit, logistic regression model building unit, decision tree model building unit, neural network model building unit, model verification and optimization unit, Early warning module: The early warning module includes an early warning indicator unit, a financial ratio indicator threshold setting unit, a market indicator threshold setting unit, a non-financial indicator threshold setting unit, an early warning method selection unit, a real-time pop-up reminder unit, an email notification unit, an SMS alarm unit, an early warning report generation unit, a risk type presentation unit, and a risk level assessment unit.

2. The enterprise financial risk early warning method based on big data according to claim 1 comprises the following steps: Step 1: Data collection: Collect internal financial data of the enterprise, including financial statements, account details, budget planning and cash flow records, etc., and capture external data, such as industry dynamics, market trends, competitors' financial performance and macroeconomic data; Step 2: Data preprocessing: clean the collected data, remove duplicate, erroneous and incomplete data, integrate data from different sources to form a unified data format, standardize the data, and unify the format and unit of the data for subsequent analysis; Step 3: Determine risk indicators: Select appropriate financial ratio indicators, such as debt-to-asset ratio, current ratio, net profit growth rate, etc., and include non-financial indicators, such as market share changes, customer satisfaction, supplier stability, etc.; Step 4: Risk assessment model construction: Build a model based on methods such as multivariate linear regression, logistic regression, decision tree or neural network, and use historical data to train and optimize the model; Step 5: Risk assessment and prediction: Input the preprocessed data into the model to assess the company’s current financial risk level and predict future financial risks; Step 6: Set warning thresholds: Determine the thresholds for different risk indicators based on the enterprise’s risk tolerance and industry characteristics; Step 7: Early warning signal generation: When the risk indicator exceeds the threshold, a corresponding early warning signal is generated; Step 8: Early warning transmission and response: Early warning signals are transmitted to relevant personnel through various means, such as real-time pop-up windows, emails, and text messages. The company management formulates response strategies based on the early warning signals and takes measures to reduce risks. Step 9: Model monitoring and updating: Continuously monitor the accuracy and effectiveness of the model, and update and optimize the model based on new data and business changes.

Images