Data mining and XBRL integration in management accounting information based on artificial intelligence

Abstract

In today’s society, the application of information technology is becoming more and more extensive. At the same time, management accounting, as an important branch of modern accounting, also ushered in new development opportunities, and the research of data mining also pays more attention to the combination of theory and practice. Therefore, data mining can provide some technical support for the implementation of strategic management accounting. Because the most important thing of management accounting informatization is to process, calculate and transmit the business information of the enterprise through the corresponding information processing platform through the use of computer technology, and provide the corresponding data to the management of the major companies in order to better analyze and make decisions and perfect the future development strategy of the enterprise, so the screening of the corresponding technology is more important in the process of management accounting informatization. Based on the development of artificial intelligence technology and the combination of data mining and XBRL technology, this paper discusses the new strategies of contemporary management accounting development.

Keywords

Artificial intelligence management accounting data mining XBRL technology

1 Introduction

Nowadays, the application of information technology is becoming more and more extensive. At the same time, management accounting, as an important part of modern accounting, has ushered in new development opportunities. Major enterprises began to widely use some advanced production technology, information model, management methods to deal with unprecedented large information groups. In this era of information economy, the mode of scale economy has gradually changed to the mode of speed economy, so information has become particularly important. Especially recently, the development of mobile Internet, Internet of things, cloud computing technology has promoted the direct growth of information, and the concept of big data has gradually entered the field of vision. Cloud computing and big data have a very profound impact on accounting, theory and practice are also very concerned about this. For example, under the huge background of cloud computing, intelligence, big data and mobile Internet, the 10th Annual meeting of Accounting Informatization in 2011 was conducted with the theme of “constantly innovating and creating a new era of accounting informatization” as the theme. It can be seen that strategic management accounting is facing the great challenge brought by big data in today’s era. How to give full play to the internal and external value contained in information has become the main research content of strategic management accounting. In order to deal with such big data, it is the main function of data mining to use special data mining technology and tools to find out the trend and law behind the data, and then to provide support for decision-making research. In the era of coexistence of knowledge and economy, in the face of the multiple challenges of big data, strategic management accounting needs to play its strategic support role in order to play a large amount of information value. In the 21st century, the research of data mining focuses more on the combination of theory and practice. Therefore, data mining can provide more powerful technical support and support for the implementation of strategic management accounting.

As an important part of the development of accounting informatization in China, management accounting informatization has been attached great importance to in the process of carrying out business. Moreover, the training and application of management accounting talents gradually began to be a major purpose and goal of enterprise development. Because management accounting is becoming more and more information, mainly by using computer technology, the business information of enterprises is processed and transmitted through the information processing platform. It is in this context that the application of XBRL is particularly important to explore the process of management accounting informatization. Based on the development of artificial intelligence technology and the combination of data mining and XBRL technology, this paper discusses the new strategies of contemporary management accounting development.

2 Related work

In terms of the impact of information technology on accounting, According to [1] literature, “in the context of information technology, Accounting as an important branch of economics, Its capacity will gradually expand, At the same time, its independence will weaken to some extent, But at the same time, the relationship between accounting and other economic management disciplines is becoming more and more obvious.” According to the literature [2] under the background of the information age, The diversification of the information needs of enterprises and countries has gone beyond the previously defined range of inherent accounting, The future of accounting information systems also needs to be reconsidered. Documentation [3] from research, The complementary role of information and communication technology and management technology and decision-making mechanism has a significant impact on enterprise income. According to literature [4] information technology is changing the contrast between accounting information and other information, The core competence of accounting has become a basic problem emphasized by accounting theory and operation. Document [5] put forward the so-called “accounting information view”, The information data of accounting should be expanded from structured to semi-structured and unstructured, In order to get better returns. According [6] the literature, In the context of information technology, Accounting informatization will face the fourth climax of development, That is to say, through the experience of management information and accounting information, Accounting informatization will usher in an era of more attention to knowledge and standards, And directly into the intelligent stage of accounting personnel.

As for the application of data management technology to management accounting, the literature [7] uses artificial intelligence neural network method to make a specific prediction analysis report on the lowest and highest consumption of new products. The prediction analysis report points out that when the ratio of the lowest value interval to the highest value interval is small, the neural network method will have better prediction ability and computational performance. Document [8] by analyzing the effect of natural language application technology and data information mining technology in transmitting and selecting information, a series of schemes for mining network information are finally formulated. So that the technology can automatically trace and analyze the purpose of enterprise brand. According to the analysis of strategic management accounting of value flow [9] literature points out that data mining is a very important aspect of the utilization of strategic management accounting information resources. Then it is suggested that countries and enterprises should establish value flow-based data warehouse. Use the warehouse to meet the further needs of enterprise management for strategic management accounting. By using data information mining technology in accounting analysis and decision-making of large database [10] literature further studies and discusses the possibility of data mining tools in accounting decision support and the feasibility of this technology. The literature [11] introduces the network information mining technology into the ETP decision management system. By controlling the consumption of purchasing cost, the risk and cost of purchasing decision are effectively reduced. Document [12] based on the information management architecture, the implementation model of the information decision management system is established. The development strategy of the financial accounting subsystem and the data mining in the management accounting subsystem are fused, and the reasonable proportion of the two is finally obtained.

On the research of XBRL technology, the literature [13] holds that the XBRL technology is to transmit the information in the financial statements to the users of other countries, understand and explain it before making decisions, and make the accounting language unified. And then effectively achieve efficient communication. Literature [14] through the use of differential analysis to study some different levels of investors, it is considered that the application of XBRL technology can enable some large investors to use their strong resource capacity to obtain greater returns. And then increase the direct information asymmetry of different levels of investors. Literature [15] study whether financial reporting using extensible business reporting language will reduce surplus during the implementation XBRL of the Securities and Exchange Commission. Marriage shows that the absolute discretionary items are significantly reduced from before and after XBRL use, which indicates that the application of XBRL restricts earnings management through the choice interval of discretionary accrual items. Through analysis, literature [16] and their teams have proved that XBRL technology can become a global data ecosystem, provide an interactive financial and business information base, and it is likely to use XBRL technology beyond the current role of disclosure and regulation. The literature [17] is based on the best case of XBRL enterprise’s internal application, combined with the current development of management accounting information and the requirements of economic market development for the reconstruction of management accounting system and report. The framework theory of XBRL technology in the application of management accounting information is put forward. Literature [18] by studying and analyzing the relevant data of Shanghai Composite Index and Shanghai A ¡¢B Stock Index, it is concluded that whether the stock index is used as the index of the stock market, the financial report in XBRL format can improve the information effectiveness of the stock market.

3 Management accounting analysis based on artificial intelligence and data mining technology

3.1 Data mining algorithm analysis

The information value refers to the average number of random events in large events. By using the difference of average information amount, the instability and irregular volatility of each large discrete information ratio in the information source are reflected. Further effective description of the specific problem of information, but also in the measurement of weight and other aspects can be used. The relevant definitions and concepts are as follows: the information value of the source is to describe the average uncertainty and fluctuation volatility of the source, that is, the range of the specific gravity difference between the average information quality and the quantity of the source. It can be measured by the average difference of a single symbolic amount of information, that is: $H (A_{le}) = (I (A_{il})) = - \sum_{l = 1}^{d} p_{l} {log}_{2} p_{l}$ (1)

For the weight of the p attribute in the attribute set: $w_{l} = \frac{H (A_{l})}{\sum_{j = 1}^{d} H (A_{j})} (l = 1, 2, . . ., d)$ (2)

Weight distance between the two data differences gik calculated using the American distance formula shown below: $d_{ij} = \sqrt{\sum_{l = 1}^{d} (x_{il} - x_{jl})^{2}}$ (3)

Concept 1: weight k nearest neighbor: refers to any object on the QS’ of weighted proportional data I, According to formula (3), the distance between I and other weights is calculated, and k specified data with the smallest distance difference ratio is obtained. The final calculation average is worth another distance difference interval formula, that is: $d_{q} = \frac{d_{min} + d_{ave}}{2}$ (4)

Algorithm description: input: data set QS, data object has n, dimension d, copy has f, nearest neighbor has k:

Output: Outlier object;

1) calculate according to formula (1)(2) to get the DS’; of weighted data set

2) for (i = 0; f > i;i++){

3) $\vec{v_{i}} = (2^{m} i / (d + 1), . . ., 2^{m} i / (d + 1)) (i \in [0, d])$ (5)

// Build a specified translation vector m a ratio

4) ${DS}^{i} = DS + \vec{v_{i}}$ (6)

// Build a copy of the dataset DPS different from the later DPS

5) $Z_{i} = ComputeZ ({DS}_{j}, n)$ (7)

// The Z values of all objects in the dataset DSi are calculated

6) $C_{i} = Candidate ({DS}_{i}, n, k)$ (8)

// Construction of weighted k nearest neighbor candidate sets;

7)}

8) $C = C_{1} U . . . {UC}_{f}$ (9)

// Collate the weighted candidate sets in all copies;

9) $D_{q} = WKNN ({DS}_{i}, C, n, d, f, k)$ (10)

// All objects are weighted k nearest neighbors;

Returns the first TOP-N objects with the largest median Dq;

Since f copies are needed in the WKNNOM algorithm and n objects exist in each copy, the spatial complexity of the algorithm is O (f n).

3.2 Genetic algorithm and improvement of artificial intelligence

Genetic algorithm is a direct evolution of biological evolution process. In the final calculation, the optimal solution is obtained. Due to the different fitness of individuals, the principle of survival of the fittest and survival of the fittest is carried out. The expression formula is as follows: $P = {\begin{matrix} P_{c 1} - \frac{(P_{c 1} - P_{c 2}) (f_{max} - f^{'})}{f_{max} - f_{avg}}, f^{'} ⩾ f_{avg} \\ P_{c 1}, f^{'} ⩾ f_{avg} \end{matrix}$ (11) $P = {\begin{matrix} \frac{(P_{c 1} - P_{c 2}) (f_{max} - f^{'})}{f_{max} - f_{avg}}, f^{'} ⩾ f_{avg} \\ P_{c 1}, f^{'} ⩾ f_{avg} \end{matrix}$ (12)

The core idea is to retain a large fitness individual within a certain distance, which can not only protect the diversity of individuals and populations, but also make a certain distance between individuals, and then avoid falling into a local optimal solution. $f^{'} (X_{i}) = f (X_{i}) / S_{i}, i = 1, 2, . . . m$ (13)

Holt-Winters model expression: $Y_{i} = α \frac{X_{i}}{H_{i - L}} + (1 - α) (Y_{i - 1} + G_{i - 1},)$ (14) $G_{i} = γ (Y_{i} + Y_{i - 1}) + (1 - γ) G_{i - 1}$ (15) $H_{i} = β \frac{X_{i}}{γ_{i}} + (1 - β) H_{i - L}$ (16)

Sales estimates can be obtained by following: $Y_{i + s} = (Y_{i} + G_{i} s) H_{i - L + s}$ (17)

The α,β and γ values obtained by the improved genetic algorithm are time-consuming and can obtain the optimal solution.

3.3 Framework of management accounting system based on data mining

According to the network information mining technology management accounting decision-making mode refers to the development of strategic management accounting activities under the background of cloud computing and big data, using the principle and technology of information data mining, and according to the overall structure of strategic management accounting of data mining, It includes basic construction layer and measurement layer, information screening and collation layer, data storage layer and so on, as shown in Figs. 1 and 2.

Fig. 1

Basic Principles of Data Mining Technology for Strategic Management Accounting.

Fig. 2

Overall Framework of Strategic Management Accounting Based on Data Mining.

Through the analysis of the application structure of information data mining in decision management accounting, it can be concluded that under the guidance of its theory and method research, the enterprise’s intelligent management strategy is the primary goal. The information flow from different data sources is analyzed and screened. Data mining and utilization technology play a certain role in the screening layer and knowledge selection layer of information. The purpose is to realize the transformation from data to information and then to wisdom, so as to improve the value of enterprises and finally improve the comprehensive competitive strength of enterprises.

3.4 Management accounting implementation process based on information mining

Decision-making process criteria (DPOIS-XM) for cross-industry large information mining focus on information mining not only for data reorganization and re-presentation, but simply for data analysis and modeling, but also a process from understanding business requirements, finding decision management to receiving screening and detection. Generally, it includes six stages: business construction, screening, decision making, modeling, evaluation and deployment, see Fig. 3 for details.

Fig. 3

General Data Mining Process.

The ultimate goal of successful data mining is to make itself commercially valuable, not just through complex tools and algorithms. The purpose of data mining is to find potential patterns in a large number of complex data and transform data into useful knowledge. Based on the principle and framework of information data mining applied to strategic management accounting, the relevant staff put forward the implementation process of big data decision application in strategic accounting management.

(1) The first is to define the problem according to the requirements of strategic management accounting, and to further determine the internal and external information based on the problems and purposes of analysis. This step can be done between corporate executives and ordinary employees, and then set topics and goals, which can make information search more systematic. Management and to put forward the analysis goal, but also through internal consultation, to avoid invalid demand and unreasonable demand, and finally to improve the ability to analyze the data, reduce the cost of consumption, and finally achieve the purpose of avoiding the waste of resources and energy.

(2) After a clear analysis of the problem, the internal and external information of the data collection system in the business management activities will be incorporated into the database. The essence of this process is to digitize the decision-making process of the enterprise manually and generate a new resource bank through the network processing of information system to provide it to the management at all levels. Therefore, relevant personnel can use the internal information decision management system to obtain internal decision management information. According to the change of decision-making management mode, enterprises can further rough the collected data, and at the same time, use the auxiliary input function of computer network to income it into the large database. The information in the big database is processed into the warehouse through a series of programs, so as to realize analysis and re-mining.

(3) These coarse-processed information, such as filtering and integration, need to be analysed to translate into available knowledge, which is also the core of strategic management accounting. At this stage, strategic management accounting includes various aspects to provide sufficient analytical ideas for the determination of data mining topics, data understanding, model selection, result analysis and evaluation; in addition, the analysis and evaluation of data mining results also need to be fed back to relevant personnel. Then prove and optimize the use of strategic analysis tools.

(4) The service system of information data mining takes the initiative and transmits information in a timely manner by analyzing the results of customer requirements or accumulating information according to the amount of information within the system. The results presented according to market analysis data can be presented through user experience, active push, fixed value and so on. Aiming at the problem of unsynchronization and differentiation of requirements, the information leakage is prevented by setting and defending user rights.

(5) Peter Drucker once said that in the age of broad knowledge economy, every employee is actually a manager. Managers at all levels of enterprises (including ordinary employees) can acquire knowledge according to their respective authority, and combine the concept and method of strategic management accounting to improve the level of business decision-making and enhance the ability and wisdom of employees.

4 Application of XBRL based on management accounting information data mining

4.1 Research methodology

This research scheme mainly adopts the event research method, its principle is according to the market validity theory, according to the research purpose principle a specific event, and collects the related data to calculate the sample stock return rate change before and after, Finally, through the change data and trend, how to perfect and update the sample stock price and return change. This method is often used to account for the degree of response of accidental events to price changes. The change of stock price is usually caused by the change of an index in the financial report of an enterprise. Because XBRL can realize the correspondence of financial index through the setting of classification standard, it can present a variety of enterprise information for users at all levels, and to some extent, it will cause stock price fluctuation. Therefore, the event research method is effective to some extent.

The event day refers to the time node in which the market accepts the event information, and what this paper is going to verify is that XBRL can effectively increase the quality of relevant financial information. So we choose the time node XBRL network financial report presented by G company as the event day.

XBRL is a new technology, its network financial report for information users need some time to adapt to more fully play its functions. For the choice of window period, it can not be too long or too short, too long will have the influence of other uncertain factors, too short can not draw a persuasive conclusion. Based on the development practice of G company and the previous hypothetical reasoning, we select the window period of [-1] to study and draw the relevant conclusions.

4.2 Model design

In order to estimate whether the XBRL financial report has a significant return on the stock price of the listed company, we should first calculate the advance price normally. The main function of this paper is the market price model method. The principle of this method is to infer the future stock price change of the enterprise by using the correlation between the market index and the previous stock price of the enterprise. Based on the previous stock price of the enterprise, the correlation between the stock price and a certain market index is calculated. For example, the BETA value to be calculated in this paper. Second, calculate the daily return rate of the enterprise stock according to the correlation index of the first step. Finally, the fair price of the company’s shares on the same day is calculated according to the value of the second step. The formulas required for this model are as follows: $R_{it} = α + β R_{mt} + ɛ$ (18) $\overset{\land}{R_{it}} = \overset{\land}{α} + \overset{\land}{β} R_{mt} + ɛ$ (19)

Based on the calculation model formula, the cumulative excess return rate is calculated by physical research method. The detailed process is as follows:

(1) Determination of dates. Since this model needs to test whether the application of XBRL technology can increase the quality of information disclosure, the event time node selected in this paper is the date of disclosure XBRL the report. The window period is [-1] and the estimated period is [-30-1].

(2) Selection of data sources. To test whether the disclosure of financial system reports obtained by XBRL technology has a significant impact on the stock price of enterprise companies, this paper selects M companies as the research object through layer by layer screening. The information quality of the financial report is analyzed more deeply by taking the financial report disclosed XBRL the company in 2010-2017. A further analysis of the relevant data sources and the company’s annual financial statements and WIND database.

(3) Calculating stock price gains and market value daily gains. Based on the closing price of stock price, the daily income of stock is further calculated, and the market closing index is used as the basis. The formula is as follows: $R_{it} = (P_{it} + P_{i, t - 1}) / P_{i, t - 1}$ (20) $R_{mt} = (P_{mt} - P_{m, t - 1}) / P_{m, t - 1}$ (21)

(4) Calculation of estimated coefficients $R_{it} = α + β R_{mt} + ɛ$ (22)

(5) Calculating the normal earnings of each stock during the screening period. The formula is as follows; $\overset{\land}{NR} = \overset{\land}{α} + R_{mt} * \overset{\land}{β} + ɛ$ (23)

The estimated value (AR) of the abnormal rate of return and the estimated value (CAR) of the cumulative excess rate of return are calculated by the following formula: $\overset{\land}{AR} = R - \overset{\land}{NR}$ (24) $\overset{\land}{CAR} = Σ \overset{\land}{NR}$ (25)

4.3 Analysis of experimental results

Budget table for major operating indicators as show in Table 1, breakdown of products and contracts as show in Table 2. Budget statement for profit and loss as show in Table 3.

Table 1
Budget table for major operating indicators (in 10,000 yuan)

2018 2018 Year-on-year growth

Operating income 8000 12000 50%

Total profits 2000 2600 30%

Cost costs 75% 78% 4%

EVA 2800 3000 7%.

Multiple of surplus cash security 3000 1.13. 6%

Net receivables 500 4000 33%

Net inventory 2016 900 80%

	2018	2018	Year-on-year growth
Operating income	8000	12000	50%
Total profits	2000	2600	30%
Cost costs	75%	78%	4%
EVA	2800	3000	7%.
Multiple of surplus cash security	3000	1.13.	6%
Net receivables	500	4000	33%
Net inventory	2016	900	80%

Table 2

Breakdown of products and contracts (in 10,000 yuan)

Year 2017		First quarter	Second quarter	Third quarter	Fourth quarter	Total
A plate	Projected	15,000	15,000	20,000	20,000	70,000
Last year	10,000	10,500	10,000	9,000	39,500
Year-on-year growth	50%	43%	100%	122%	77%
BPlate plate	Projected	103	103	137	107	50
Last year	300	320	300	270	1,190
Year-on-year growth	–66%	–58%	–54%	–49%	–60%
CPlate plate	Projected	1,000	900	000	1,000	3,900
Last year	1,100	1,500	500	2,200	6,300.
Year-on-year growth	–9%	–41%	–33%	–55%	–38%
DPlate plate	Projected	1,000	1,500	2,394	2,756	7,650
Last year				510	510
Year-on-year growth
Total income	2103	2503	3531	3863	12,000

Table 3

Budget statement for profit and loss (in 10,000 yuan)

Project	Last year	Budget for the year	Increase or decrease (%)
I. Gross operating income	8,000	12,000	50.00%
II. Total operating costs	5,870	10,170	73.25%
Among them: main business cost	400	5,400	285.71%
Operating Tax and Additional	50	78	56.00%
Sales expenses	500	2,500	66.67%
Management costs	2953	2232	–24.43%
Financial costs	–33	–40	19.01%
Loss of assets	200	230	15.00%
III. Operating profits	930	1600	–17.10%
Plus: non-operating income	100	1000	900.00%
Less: Non-operating expenses	30	2,600	–100.00%
IV. Total profits	2,000	30.00%
Less: Income tax expenses	500	650	30.00%

Cost budget as show in Table 4. Evaluation of Business Quality as show in Table 5.

Table 4

Cost budget (in 10,000 yuan)

Project	2018	2019	Last year
(i) Sales expenses	1500	2,500	67%
of which 1. Transportation costs	8	10	25%
2. advertising	2	350	17400%
3. Land Pay	500	900	80%
4. operational requirements	260	300	15%
(ii) Administrative expenses	2953	2232	-24%
Of which 1. Employee compensation	112.13	181.6	62%
2. business hospitality	5	10	100%
3. travel	8	8	0%
4. office expenses	16	18	13%
5. bus fare	10	16	60%
6. training	30	50	67%
7. rental	18	18	0%
8. research and development costs	1000	1500	50%
(iii) Financial expenses	–33	–40	18%
Total	4,419	4,692	6%

Table 5

Evaluation of Business Quality

	Basic target values	Expected target value	Compliance with criteria
Accounts receivable	The proportion of operating income fell by 3% year on year	5000	Yes
Inventory	The proportion of operating income decreased by 5% year on year	900	Yes
Number of personnel	Increase in productivity	205	Yes
Per capita value output	The deviation from the total per capita profit index should not be too large	12	Yes
Total wages	Growth rate not higher than total profit growth rate	2108	Yes
Operating cash flows	Positive	288	Yes

As can be seen from Figs. 4 and 5 above, Combined with 2019 comprehensive budget data, the cost-benefit model of basic business and value-added services is established, see Table 6 for details.

Fig. 4

Budget statement for profit and loss (in 10,000 yuan).

Fig. 5

Cost budget (in 10,000 yuan).

Table 6

Table of Quantitative Profits Model (in Meta)

	Basic business			Value-added services
	Terminal sales	Service charge	Insurance premium	Refueling value	Logistics value added	Total
Quantity	15,000	325,000	80,000	44,800	80,033	——
Sales unit	300	120	20	600	600	——
Sales revenue	4,500,000	39,000,000	1,600,000	26,880,000	48,020,000	120,000,000
Purchase unit	280	120	120	——
Variable costs	4,200,000	32,500,000	5,376,000	9,603,960	51,679,960
Fixed costs	1,500,000	400,000	420,040	2,320,040
Gross	–1,200,000	6,500,000	1,600,000	21,104,000	37,996,000	66,000,000

4.3.1 Model estimates

Since M company is a listed company on the Shenzhen Stock Exchange, the market model is used to calculate the normal income of individual stocks. The above rate of return represents the size of the overall market income, and the first 30 trading dates are disclosed in the annual XBRL financial report. By using the formula, the annual market adjustment model is estimated OLS (least square method), and the results are as shown in Table 7:

Table 7
Market adjustment model estimates

Model R of dependent variablesit

Unstandardized coefficient Standardized coefficients Significant significance

B Standard error Beta t

2013 (Constants) 0.058 0.149 — 0.391 0.699

Rmt 0.542 0.163 0.532 3.322 0.002

2014 (Constants) 0.220 0.471 — 0.467 0.644

Rmt 1.215 0.431 0.405 2.822 0.009

2015. (Constants) 0.220 0.471 — 0.467 0.644

Rmt 1.215 0.431 0.471 2.822 0.009

2016 (Constants) –0.143 0.261 — –0.548 0.588

Rmt 1.165 0.209 0.725 5.577 0.000

2017 (Constants) –0.139 0.225 — –0.619 0.541

Rmt 1.200 0.121 0.883 9.958 0.000

2018 (Constants) –0.090 0.180 — –0.502 0.619

Rmt 0.896 0.343 0.443 2.613 0.014

2019 (Constants) 0.235 0.269 — 0.873 0.390

Rmt 1.191 0.212 0.728 5.614 0.000

Model	R of dependent variablesit
2013	(Constants)	0.058	0.149	—	0.391	0.699
Rmt	0.542	0.163	0.532	3.322	0.002
2014	(Constants)	0.220	0.471	—	0.467	0.644
Rmt	1.215	0.431	0.405	2.822	0.009
2015.	(Constants)	0.220	0.471	—	0.467	0.644
Rmt	1.215	0.431	0.471	2.822	0.009
2016	(Constants)	–0.143	0.261	—	–0.548	0.588
Rmt	1.165	0.209	0.725	5.577	0.000
2017	(Constants)	–0.139	0.225	—	–0.619	0.541
Rmt	1.200	0.121	0.883	9.958	0.000
2018	(Constants)	–0.090	0.180	—	–0.502	0.619
Rmt	0.896	0.343	0.443	2.613	0.014
2019	(Constants)	0.235	0.269	—	0.873	0.390
Rmt	1.191	0.212	0.728	5.614	0.000

Table 7 reports the statistical characteristics of the AR of M companies before and after the disclosure of their financial reports. According to the results, we can see that the average AR before disclosure is –0.089, and the average AR after disclosure reaches 0.0125, an increase of nearly 0.2 percentage points. To some extent XBRL the abnormal rate of return of stocks after financial report disclosure has increased significantly, see Table 8 for details.

Table 8

XBRL (%) of statistical features before and AR financial reporting

Project		G Company 7
XBRL Prior to financial	Average value	–0.08691
disclosure	Standard deviation	0.978731
XBRL disclosure of	Average value	0.121377
financial reports
Standard deviation	1.731077

Using the empirical analysis framework of event research, we quantitatively analyze the size and changes of the CAR and AR before and after the financial reports of M companies from 2010 to 2017. We can see that the trend of CAR changes shows a significant upward trend, and the average value of the AR is significantly higher than before the disclosure of the financial reports, which indicates that XBRL financial reports improve the level of AR, Furthermore, it proves that XBRL financial report has a very positive impact on the quality of information. To some extent, the use of XBRL technology can improve the quality of management accountants, so as to better advise the management of major enterprises.

5 Fully explore the application value of XBRL in management accounting information system

5.1 Establishment of a dynamic monitoring system for competitors in the same industry

At present, the financial report presented on the XBRL platform of Shenzhen Stock Exchange can compare the data of five peer enterprises at the same time, which provides a very obvious advantage for the development of enterprises at this stage. At the same time, the Ministry of Railways has organized and developed the RFAMIS (Railway Financial Accounting Management Information system), which is a syndrome system that uses computer technology to provide financial accounting and management decision-making services for railway financial workers. If the M company can combine the management accounting information system with the RFAMIS, it can design a set of competitive strength for the management accountant, and monitor the dynamic of the opponent. Through this system, M company can compile the corresponding report of monitoring industry opponent according to the management accountant, which will also become a new basis for the enterprise management to make the discovery decision.

Moreover, in order to maximize the value of such monitoring systems, it is not enough to rely on the efforts of M companies or G companies alone. A similar monitoring system organization can be set up with the leadership of official government authorities, such as the financial sector, in which enterprises of the same industry are grouped into a group, with members holding regular academic exchange seminars to share the success of the application of new technologies, and members cooperating and making progress together in order to continuously improve the monitoring system and finally realize the virtuous circle of competition between enterprises.

5.2 Development XBRL application in performance management informatization

As one of the basic functions of management accounting, performance management also plays a very important role in the development of management accounting informatization. By referring to the current situation of enterprises that have implemented budget management informatization at this stage, we can see that although the existing information systems can solve a series of traditional problems, such as large amount of evaluation data collection and difficult application of evaluation results, there are still some limitations in the application process due to the imperfect budget system and complex operation. According to the technical advantages of the XBRL and the application status of the M company, if this is applied to the information process of performance management, these defects can be made up to a great extent.

First of all, because performance management involves the whole business situation from top to bottom and the working situation of each employee, the characteristics of XBRL technology can be used to further develop its classification criteria, so that the technology can cover every employee, every business, and can be presented in a more detailed summary at the end of the month, quarter and year, thus providing some convenience in the assessment and analysis of management accountants.

Secondly, enterprises can cooperate and communicate with the suppliers of financial software according to their own performance evaluation demands, jointly develop the operating software for the application of XBRL technology, and fully reflect the advantages of openness and transparency of XBRL information. In order to facilitate management accountants to collect and summarize all the data.

Finally, the use of XBRL technology in the company’s information system to all members of the company can clearly show personal work plan and job responsibilities, departmental business indicators and a series of performance evaluation content, can more intuitively promote the improvement and promotion of corporate performance.

6 Conclusion

As a new financial information exchange technology, XBRL not only improves the efficiency of management accounting, but also promotes the development of management accounting information, but also provides a very strong technical support for the long-term development of enterprises. Application of XBRL technology in the process of management accounting information is not only a simple integration, but also requires the cooperation of various teams, including enterprise management, grass-roots management accounting staff and financial software programmers. At the same time, it also needs continuous improvement in practice. At present, although XBRL technology has been used for many years, its technical advantages have not been fully excavated and applied. Therefore, it is hoped that in the process of future research and practice, the relevant staff can develop and explore the more potential functions of XBRL technology, so that it can better serve the major enterprises.

Footnotes

Acknowledgment

Soft science research plan project of Henan province, Item number: 202400410426.

References

Kumaran

and Bhavani

, MRI image retrieval using Gabor wavelet based texture features, Int J Adv Res Sci Technol 2(1) (2013), 2320–1126.

Liu

Y-H.

, Muftah

, Das

, Bai

Li.

, Robson

and Auer

, Classification of MR tumor images based on Gabor wavelet analysis, J Med Biol Eng 32(1) (2012), 22–28.

Mahajan

G.S.

and Bhagat

K.S.

, Survey on medical image segmentation using enhanced K-means and kernelized fuzzy C-means, Int J Adv Eng Technol 6(6) (2014), 2531–2536.

Malviya

M.A.M.

and Joshi

AS.

, Review on automatic brain tumor detection based on Gabor wavelet, Int J Eng 3(1) (2014), 53–56.

Manikandan

, Monolisa

G.S.

and Saranya

, A cluster based segmentation of magnetic resonance images for brain tumor detection, Middle-East J Sci Res 14(5) (2013), 669–672.

Mohan

and Subashini

M.M.

, MRI based medical image analysis: survey on brain tumor grade classification, Biomed Signal Process Control 39(3) (2018), 139–161.

Moslehi

and Haeri

, An evolutionary computation-based approach for feature selection, J Ambient Intell Hum Comput 19(6) (2019), 1590–1624.

Nabizadeh

and Kubat

, Brain tumors detection and segmentation in MR images: Gabor wavelet vs. statistical features, Comput Electr Eng 45(6) (2015), 286–301.

Noureen

and Hassan

, Brain tumor detection using histogram thresholding to get the threshold point, IOSR J Electr Electron Eng 9(5) (2014), 14–19.

10.

Patel

and Doshi

, A study of segmentation methods for detection of tumor in brain MRI, Adv Electron Electric Eng 4(3) (2014), 279–284.

11.

Potdukhe

and Nagtode

S.A.

, Application of wavelet transform for brain tumour detection, Int J Innov Res Comput Commun Eng 4(3) (2016), 3025–3030.

12.

Raja

N.S.M.

, Fernandes

, Dey

, Satapathy

S.C.

and Rajinikanth

, Contrast enhanced medical MRI evaluation using Tsallis entropy and region growing segmentation, J Ambient Intell Hum Comput 18(8) (2018), 689–702.

13.

Ruchita

R.A.

, Suresh

and Salankar

, A review on brain MRI image segmentation clustering algorithm, IOSR J Electron Commun Eng (IOSR-JECE) 11(1) (2016), 80–84.

14.

Sauwen

, Acou

, Van Cauter

, Sima

D.M.

, Veraart

, Frederik Maes

, Himmelreich

E.A.

and Van Huffel

, Comparison of unsupervised classification methods for brain tumor segmentation using multi-parametric MRI, Neuroimage Clin 12(4) (2016), 753–764.

15.

Selvapandian

and Manivannan

, Performance analysis of meningioma brain tumor classifications based on gradient boosting classifier, Proc Int J Imaging Syst Technol 28(8) (2018), 1–7.

16.

Sharif

, Tanvir

, Munir

E.U.

, Khan

M.A.

and Yasmin

, Brain tumor segmentation and classification by improved binomial thresholding and multi-features selection, J Ambient Intell Hum Comput 75(3) (2018), 1075–1083.

17.

Shingade

and Jain

, Identification of brain tumor using Gabor wavelets, Techniques 4(2) (2017), 16–21.

18.

Singh

, Arora

and Sinha

, Survey on MR image segmentation using fuzzy C-Means algorithm, Int J Eng Res 6(5) (2017), 261–264.

19.

Soltaninejad

, Yang

, Lambrou

, Allinson

, Jones

T.L.

, Barrick

T.R.

, Howe

F.A.

and Ye

, Supervised learning based multimodal MRI brain tumour segmentation using texture features from supervoxels, Comput Methods Progr Biomed 15(7) (2018), 69–84.

20.

Verma

, Tiwari

and Rahim

, Unsupervised MRI brain tumor detection techniques with morphological operations, Int J Sci Eng Technol Res (IJSETR) 4(11) (2015), 3710–3715.

21.

Andersson

, Donalek

, Farmer

, et al., Causes of the major grid blackouts in north america and europe, and recommendedmeans to improve system dynamic performance, IEEE Trans Power Syst 20(5) (2005), 1922–1928.

22.

Balaraman

and Kamaraj

, Transmission congestion management using p swarm optimization, J Electr Syst 7(1) (2011), 54–70.

23.

Bhattacharjee

and Chakraborty

A.K.

, NSGAII-based congestion management in a pool-based electricity market incorporating voltage and transient stability, Electric Power Comp Syst 41(10) (2013), 990–1001.

24.

Carlos

C.A.C.

, Evolutionary multi-objective optimization: some current research trends and topics that remain to be explored, Front Comput Sci China 3(6) (2009), 18–30.

25.

Christie

, Wollenberg

and Wangensteen

, Transmission management in the deregulated environment, Proc IEEE 88(3) (2000), 170–195.

26.

Coello

C.A.C.

, Pulido

G.T.

and Lechuga

M.S.

, Handling multiple objectives with p swarm optimization, IEEE Trans Evolut Comput 8(3) (2004), 256–279.

27.

Conejo

A.J.

, Milano

and García-Bertrand

, Congestion management ensuring voltage stability, IEEE Trans Power Syst 21(1) (2006), 357–364.

28.

David

A.K.

and David

, Dynamic security enhancement in power-market systems, IEEE Trans Power Syst 17(2) (2002), 431–438.

29.

Dutta

and Singh

S.P.

, Optimal rescheduling of generators for congestion management based on p swarm optimization, IEEE Trans Power Syst 23(4) (2008), 1560–1569.

30.

Esmaili

, Shayanfar

H.A.

and Amjady

, Congestion management considering voltage security of power systems, Energy Convers Manage 50(1) (2009), 2562–2569.

31.

Esmaili

, Shayanfar

H.A.

and Amjady

, Multi-objective congestion management incorporating voltage and transient stabilities, Energy 34(2) (2009), 1401–1412.

Model		R of dependent variablesit
	Unstandardized coefficient		Standardized coefficients	Significant significance
	B	Standard error	Beta	t
2013	(Constants)	0.058	0.149	—	0.391	0.699
Rmt	0.542	0.163	0.532	3.322	0.002
2014	(Constants)	0.220	0.471	—	0.467	0.644
Rmt	1.215	0.431	0.405	2.822	0.009
2015.	(Constants)	0.220	0.471	—	0.467	0.644
Rmt	1.215	0.431	0.471	2.822	0.009
2016	(Constants)	–0.143	0.261	—	–0.548	0.588
Rmt	1.165	0.209	0.725	5.577	0.000
2017	(Constants)	–0.139	0.225	—	–0.619	0.541
Rmt	1.200	0.121	0.883	9.958	0.000
2018	(Constants)	–0.090	0.180	—	–0.502	0.619
Rmt	0.896	0.343	0.443	2.613	0.014
2019	(Constants)	0.235	0.269	—	0.873	0.390
Rmt	1.191	0.212	0.728	5.614	0.000