Printer Friendly

Research on the internet financial accurate model for poverty alleviation based on multi-agent and data mining.

1. Introduction

Since 2014, the China's macro-economic development started to enter the high efficiency, low cost, rapid growth in sustainable state, known as the "new normal" stage (Fanadzo, 2012; Crespo, 2015). The new normal is a time in the future economic and social development to run the fundamental characteristics of the "Internet+" is the new normal driver development is an important part of the innovation; innovative industry will face unprecedented rapid development opportunities (Alix, 2013; Burney, 2012). And as to interest marketization accelerating and private Banks of gradually introduced a series of major financial reform, gradually formed a future financial development of the new normal (Faust, 2012; Francis, 2013; Banson, 2014). Simply put, the Internet financial is the way of all use the Internet to provide financial services. At present, for the complicated Internet financial mode has not unified standard of classification, we introduce a new classification basis, namely: (1) the pattern of the extent of the application of Internet technology and the Internet on the depth of the influence of the model; (2) the model of information production way. In the table one, we demonstrate the characteristics of them, respectively.

The financial constraints of the new normal Internet financial need to integrate the information technology into the service means, innovative business model, efficient service the real economy and achieve development with their own wisdom and connotative growth (Fisher, 2014; Khanna, 2015). The Internet finance and traditional finance there are both commonalities and differences. Common starting point is that Internet financial still around to meet customer payment, deposit, the basic requirements, such as financing and financing just implanted into the Internet "gene". The difference lies in the following aspects: first is the difference channel. Internet financial mainly relying on online acquisition channels, while traditional financial main position in line. The second is the use of big data. Accordingly, in the figure one, we show the general rank of the reputed finance websites.

Technological progress to reduce the cost of trading and the traditional financial industry financing demand for the mid-range indifference and gave the Internet financial with great population, directly led to the financial the germination and growth of the Internet, a profound influence on the formation of financial forms. As the abundance of the Internet financial products, and users of the financial habits foster, customer's perception of service convenience, yield gradually deepened, to risk identification are also gradually increase (Rai, 2014). Online interactive development of financial services making rapid progress in industry, Internet thinking permeates into the financial sector. This disruptive service oriented, has made the traditional financial institutions under pressure, transformation is imminent. Institutions such as the banks, insurance, securities firms began to adjust the strategy, the Internet technology, to adapt to the new financial system. Under this historical condition, in this paper, we conduct research on the Internet financial accurate model for the poverty alleviation based on the multiagent and data mining. The detailed analysis and discussion will be given in the later sections. To begin with, we illustrate the developmental trend of the Internet financial methods in the figure two.


2. The data mining technique and financial systems

Data mining is widely used in all fields at present. Such as on the geographical database mining geological and geomorphic characteristics of looking for mineral or urban planning; On a Web server Web log mining, according to user's interest in dynamic link Web page, the statistics page links, statistical authority home page, etc., and can be carried out on the search page clustering, easy to find the needed information (Trau, 2012; Urge, 2012). For biomedical DNA data analysis and mining data characteristics such as heredity, diseases; Analysis of the financial data mining, customer credit; Mining, guidance to the data in the retail shelves and goods order. Based on the review, the categories of the data mining can be listed as follows.

1. Hierarchy model based on microscopic characteristics of the data found its characterization, with universality, the concept of the higher level, the knowledge of the macro that reflect similar things common properties, a generalization of the data, refining and abstract.

2. Classification mode reflects the characteristics of similar things common nature of the knowledge and the differences between the different things characteristics of knowledge.

3. Sequential patterns, given a set of different sequences. Among them, the orderly arrangement of each sequence made of different elements, each element is composed of different projects. At the same time, given a user to specify a minimum support rich value, sequence pattern mining is to find all the frequent subsequence, namely the subsequence in sequence of the frequency of not less than the user to specify minimum support threshold.

4. Deviation mode, the exception is in the data set different data, make people suspect that the data is not a random deviation, on the completely different mechanisms but most of the data mining method to the data as a noisy and cast it away.

As for the data mining application in the finance area, we firstly analyze the basic high-frequency financial analysis model. Financial high frequency data sequence usually refers to the frequency of hours, minutes or seconds, according to the time sequence of the collection of financial data (Zhang, 2015; Yusuf, 2016). In general, the sampling frequency is proportional to the market information, the higher the sampling frequency, the more the market information is obtained; otherwise, the less. According to the different sampling methods, it can be divided into sub-time data and sub-data (Turner, 2012). In calculating VaR aspects of financial assets, people are constantly developing new copulas connect model to describe the dependence structure of multiple portfolio. Such as Bedford and Cooke in simple structure module pair copulas connect on the basis of introduced a new method of the multiple correlation structure model, the complex structure it will multivariate probability density function is decomposed into a series of pair copulas connect modules and the edge density function of the product, it's for copulas connect method is extended to the higher dimensional case provides the theoretical basis. For the high dimensional joint distribution, the pair copulas connect down there are many kinds of the logical structure. The following figure three illustrates the architecture.


For this logic, we could use the model shown in formula one to represent.

F(x|v) = [partial derivative]C(F(x | [v.sub.-j]),F(v, | [v.sub.-j]))/[partial derivative]F ([v.sub.j] | [v.sub.-j]) (1)

In the multi-asset portfolio risk analysis, it is not enough to know a single asset marginal distributions, the dependency structure between the assets must also be considered. Although n copulas connect function can describe the dependency structure between variables, but it ignores the influence of the dimension and the tail correlation differences between the two assets, pair of the multivariate joint probability density function decomposition of basic copulas connect can solve this problem. The distribution can be transferred into the formula 2.


In addition to the Gaussian copulas connects, they in the description of the binary variable distribution on the tail correlation each are not identical. Such as Clayton copulas connect more suitable for fitting with only the tail correlation data, and Gumbel copulas connect, on the other hand, the copulas connect with upper and lower tail correlation data at the same time for fitting. In the empirical analysis, based on the original data and conditions observations of scatter plot or test of goodness of fit (GOF test) to determine which pair picking function types of data fitting is best. The further steps for analysis of the issues will be listed as follows.

1. Cross-validation technique is used to determine a prediction model for the required number of the principal components and build predictive function parameters.

2. Principal component analysis of the historical value and gain characteristic function type data object, data object and the average function and limit respectively two data objects in a given breakpoint range.

3. Sampling data from flow every once in a fixed length of the time to set the breakpoints, between adjacent breakpoints set a variable, the breakpoint between the sequence of values for the variables corresponding to a series of observations.

In ARMA model predictions considering the financial time series has short-term correlation, all variables in this are no longer used as the prediction data fitting model, this article takes 35 top 20 value of a variable modeling data to make the predictions. Because of the ARMA model is more suitable for stationary time series, and financial time series are often reflects the certain stability need to be smooth. In the figure 4, we show the visualized procedures of the modelling.


3. Multi-agent model and the financial systems

Information resource collection system based on Multi-Agent by the Agent for the information retrieval, scheduling and processing, can be distributed information management, control, and that classification. System is mainly controlled by the management Agent, the Agent information collection, analysis, the Agent of Agent, information management and information database. Including the information collection Agent is the key to the system, this is because the information in the Intemet information space resources are heterogeneous, and the information is a dynamic change, people want to have now, collect and maintain the information you need to be users spend a lot of time and energy. From the systematic angle, the systematic architecture of the Multi-agent model can be organized as follows.

1. Information collection system's main function is to help users such as the intelligent agent with the fastest speed with the issues related to data. As intelligent information collection system, on the whole are subject to the user or the agents such as scheduling, completed its mission requirements.

2. Management control Agent based on regular users' information demand, command scheduling of each Agent in the system operation, maintenance and management of all the communication and collaboration, the Agent in the Agent between resource conflict or contradiction to resolve, it obey the user task instructions, and information timely feedback to the user.

3. Run under the environment of network information collection Agent, using mobile Agent technology to traverse the Web, as much as possible to collect all kinds of information resources. When the activation conditions have been met, the Agent of perceptron receives the external information flow, according to received information processor communication primitives, communication knowledge, the current state of the control rules and its own decide whether to accept the task, and activate corresponding parts.

4. Information management is the function of the Agent will analyse filtering Agent to sort out the standardization of the information resources committed to the database, and information resource in the information to add, delete, modify, sorting, and query management, etc.


As shown above, the architecture of the multi-agent system. Due to the agent's ability and resource constraints, the agent needs to cooperate with each other to try to accomplish the aim of the system. Therefore, in order to build an effective MAS, requires the synergistic process in the system is analyzed and designed in detail, through agent collaboration to study and solve complex optimization problems. At present, the key methods of solving the problem of coordination in MAS are as follows. (1) Contract net agreement is the most famous cooperative contract. The market mechanism is adopted to achieve the effective cooperation through the task sharing, and the introduction of the non-compliance of the punishment measures, the use of the financial options pricing theory to obtain the flexible coordination scheme in the uncertain environment. (2) Abstract the collaborative problem to planning problem, as the existing centralized and distributed two Agents planning model planning need node sharing and processing large amounts of information, so involves the more computing and the communications. (3) Take the organization, hierarchical organization structure to provide a systemic view of principal Agent, to solve the conflict of each Agent by it, to ensure a uniform system behavior.


4. Internet financial accurate model for poverty alleviation

4.1. Review analysis of the Internet finance

Compared with traditional financial, Internet financial transparency is stronger, more participation, more rapid operation, intermediate use of lower cost, expand the scope of the financial sector, enrich the financial business, greatly promoted the development of the financial industry. Although in essence, the Internet financial still belong to the category of financial has inherent fragility and instability of the financial but Internet financial indeed and there is a huge difference between the traditional financial. Under this basis, we summarize the features as follows.

1. The core of the spirit of the Internet is the equal, open and sharing, the freedom to choose, and decentralization. Finance not only embodies the spirit of the Internet, the Internet also reflected the general public and the platform model for financial participation. The past can be done only by intermediary financial transactions with the help of the Internet instant platform can complete the transaction.

2. Internet banking business is based on the computer technology, network technology, information technology, science and technology as the support, can quickly to users of financial products, such as calculation, deal with the financial related business, encourage financial transactions and financial business fast and conveniently.

3. Internet banking to reduce the consumer financial integrated cost and the institutions operating costs; secondly, due to the use of the basic Internet financial level division mode, Internet financial institutions without the complicated organization, decision of flexible and effective, fundamentally overcomes the low efficiency problem of the traditional financial.

4. The effective coverage of current network technology to promote Internet banking can be smoothly carried out under the network environment, the user can not subject to the geographical and the time constraints, anytime, anywhere to financial product inquiry, purchase behavior, which will fully show that Internet banking broader coverage.

4.2. Targeted poverty alleviation suggestions

With funding and poor as the poverty alleviation and development of the specific recipients, this is the basic goal to solve the problem of rural poverty population, clothing and the basic rural operating system in our country at present stage. Poverty alleviation and development task is to solve the problem of the food and clothing of poor, only to specific funding, poor support, has it been possible to solve the problem of food and clothing of them; Rural household contract management is the basic form, only the support measures combined with the vitality of family management, to enhance the general ability of the poor self-accumulation and self-development, and then out of the poverty to get rich. Any society is impossible to completely eliminate poverty or poverty. Because poverty is an absolute concept, it is a relative concept. Society there is always a certain amount or proportion of the population is of relatively poor state. In the task of building a well-off society in an all-round way, the main task of the anti-poverty in China or in absolute poverty, to the lowest income people income or living standards improve above the poverty line, eliminating absolute poverty. Poverty alleviation and development, such as productive against poverty is still a main job, but for those who do lose labor ability and production life individuals or families in need of help, you must exert good out of social assistance system and the social security system safeguard function. Under this condition, we propose the listed suggestions for reference.

1. Encourage and support a few live in poor living conditions, lack of natural resources region extremely poor population through the way of migration, a long-distance development, open up a new way of adequate food and the clothing. Government built immigration resettlement development base, both to ensure the stable solve the problem of food and clothing of ingoing households, and to ensure that does not destroy the ingoing ground ecological environment.

2. Encourage and support the poor areas to carry out organized, planned labor export, realizing poverty-stricken areas of labor force employment. Poor area labor implementation beyond the employment, which not only helps to solve the problem of food and clothing of the poor, but also more importantly labour by different employed can learn new technology, new lifestyle and new working methods, and enhance confidence, improve the ability of self-development.

3. Weak infrastructure in poor areas, poor ability to withstand the natural disasters of the actual situation of general arrangement work-relief funds, encourages and supports poor work-relief funds, farmland and the water conservancy, roads and other infrastructure construction, improve the production conditions; The poverty alleviation and the development and resource protection, ecological construction, family planning work, realize the benign circulation of resources, environment and population, improve the ability of poor area sustainable development.

4. Focus on development of less investment, quick effect, wide coverage, high efficiency, which the directly related to resolve the problem of food and clothing planting and processing industry. Actively developing can give full play to the resource advantage in poverty-stricken areas, but also a large number of employment for poor labor resources development of township enterprises and labor-intensive.

This paper focuses on the implementation of accurate major policy for poverty alleviation of the new situation and new requirements, based on the complete "the implementation of precision, precision of poverty for poverty alleviation" major policies and measures of the state council to carry out the investigation and research situation of the third party assessment tasks, the key examines the basic characters of rural poverty in the new period, reveal the regional differentiation characteristics of rural poverty, explore the development of precision of scientific system and poverty reduction strategy for poverty alleviation, for pushing China's precision, precision of poverty a major planning and strategy for poverty alleviation and effectiveness evaluation to provide the reference.


5. Conclusion

This paper proposes the novel perspective on the Internet financial accurate model for poverty alleviation based on the multi-agent and data mining. Rural poverty alleviation work with public goods attribute, is of great is the external economic effects. Because most poor people live in resource-poor areas, in order to profit as the objective function of the enterprise has no incentive to poverty alleviation, even if corporate investment is not likely to continue for a long time, because the market of the enterprise to survive as the first priority. As corporate social responsibility increases, of course, there are quite a few enterprises through donations to non-profit institutions and thus indirectly participate in poverty alleviation is a way of industry regurgitation feeding agriculture. Under this condition, this research paper combines the multi-agent and data mining techniques to propose the new methodology of the poverty alleviation. In the future research, more related survey will be deployed to verify the effectiveness.

Recebido/Submission: 25/04/2016

Aceitacao/Acceptance: 16/05/2016


Alix-Garcia, J., McIntosh, C., Sims, K. R., & Welch, J. R. (2013). The ecological footprint of poverty alleviation: evidence from Mexico's Oportunidades program. Review of Economics and Statistics, 95(2), 417-435.

Burney, J. A., & Naylor, R. L. (2012). Smallholder irrigation as a poverty alleviation tool in sub-Saharan Africa. World Development, 40(1), 110-123.

Banson, K. E., Asare, D., Heng, L. (2014). Impact of small scale irrigation technologies on poverty alleviation among peri-urban and urban farmers. Journal of Life Sciences, 8(2), 55-60.

Betrisey, F., Mager, C., & Rist, S. (2016). Local views and structural determinants of poverty alleviation through payments for environmental services: Bolivian insights. World Development Perspectives, 1, 6-11.

Crespo, P., Santos, V. (2015). Construction of Integrated Business Management Systems for Micro and Small Enterprises. RISTI--Revista Iberica de Sistemas e Tecnologias de Informacao, (15), 35-49.

Francis, A., Nassar, A., & Mehta, K. (2013). Are we formal yet? The evolving role of informal lending mechanisms to support entrepreneurship and poverty alleviation in central Kenya. International Journal of Social Entrepreneurship and Innovation, 2(2), 109-129.

Fanadzo, M. (2012). Revitalisation of smallholder irrigation schemes for poverty alleviation and household food security in South Africa: A review. African journal of agricultural research, 7(13), 1956-1969.

Faust, J., Leiderer, S. (2012). Financing poverty alleviation vs. promoting democracy? Multi-donor budget support in Zambia. Democratization, 19(3), 438-464.

Fisher, A., Patenaude, G. (2014). Understanding the relationships between ecosystem services and poverty alleviation: A conceptual framework. Ecosystem services, 7, 34-45.

Khanna, M., Kochhar, N., & Palaniswamy, N. (2015). A retrospective impact evaluation of the Tamil Nadu empowerment and poverty alleviation (Pudhu Vaazhvu) project. The Journal of Development Studies, 51(9), 1210-1223.

Rai, R. B., Dhama, K., Chakraborty, S., Verma, A. K., Saminathan, M., Tiwari, R., & Damodaran, T. (2014). A concept of novel technological approaches for livelihood security and poverty alleviation for poor farmers: a review. Research Opinions in Animal & Veterinary Sciences, 4(7), 15-20.

Trau, A. M. (2012). Beyond pro-poor tourism: (re) interpreting tourism-based approaches to poverty alleviation in Vanuatu. Tourism Planning & Development, 9(2), 149-164.

Turner, W. R., Brandon, K. (2012). Global biodiversity conservation and the alleviation of poverty. BioScience, 62(1), 85-92.

Urge-Vorsatz, D., & Herrero, S. T. (2012). Building synergies between climate change mitigation and energy poverty alleviation. Energy Policy, 49, 83-90.

Von Maltitz, G., Gasparatos, A., Fabricius, C., Morris, A., & Willis, K. (2016). Jatropha cultivation in Malawi and Mozambique: impact on ecosystem services, local human well-being, and poverty alleviation. Ecology and Society, 21(3), 59-70.

Yusuf, M. B. O., Shirazi, N. S., & Mat Ghani, G. (2016). An empirical analysis of factors that determine poverty among the beneficiaries of Pakistan Poverty Alleviation Fund. Journal of Enterprising Communities: People and Places in the Global Economy, 10(3), 42-45.

Zhang, Q., & Chen, W. W. (2015). Multidimensional dynamic evaluation of the poverty-alleviation achievements in China's fourteen destitute areas based on GRA. Journal of Yunnan Minzu University (Social Sciences), 32(1), 136-142.

Dapeng Liu (1), Ye Wu (2), *


(1) Dianchi College of Yunnan University, Yunnan, China

(2) Yunnan Minzu University, Yunnan, China
Table 1--The Internet financial connotation and characteristics

Type            Definition

The Internet    Using the Internet technology, original
small micro     complete financial transactions
finance.        services through offline way to
                complete online via the Internet.

Pure            Due to the emergence and development of
Internet        the Internet ecosystem, formed the
financial.      financial needs of the economic
                activities on the Internet.

Internet        Represented by the counter of the
financial       traditional channels to complete the
channels.       service.

Type            Features                         Represents

The Internet    Not only the channel             P2P
small micro     alternative, and it is the
finance.        traditional financial
                services all the process of
                the implementation of the

Pure            It is entirely due to the        Third-party payment
Internet        generation of the Internet
financial.      ecosystem, the development of
                the Internet to its biggest

Internet        Traditional financial pattern    Mobile bank
financial       in a normal extension of the
channels.       Internet era, and to some
                extent, has realized the
                alternative to traditional

Figure 1--The general rank of the reputed
finance websites

Top 10 IT/Internet-Based
Business/Finance Websites

by us Market share of visits (%)

PayPal            42.5%
Interlius         11.0%
Bill Me Later      3.0%
ooVoo              2.2%
Aucativa           1.5%
OpinionLab         1.4%
MyCheckFree        1.4%
Plimus             1.2%
HomeCU             0.8%
Googie checkout    0.8%

Note: Table made from bar graph.
COPYRIGHT 2016 AISTI (Iberian Association for Information Systems and Technologies)
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2016 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Liu, Dapeng; Wu, Ye
Publication:RISTI (Revista Iberica de Sistemas e Tecnologias de Informacao)
Date:Aug 1, 2016
Previous Article:Dynamic analysis of psychological capital impact on individual growth of college students based on data mining.
Next Article:Effect of sol-gel composite treatment on corrosion resistance of magnesium alloy based on computer simulation.

Terms of use | Privacy policy | Copyright © 2022 Farlex, Inc. | Feedback | For webmasters |