Skip to main content

Advertisement

SpringerLink
Log in

Neural network rule extraction for gaining insight into the characteristics of poverty

  • Original Article
  • Published:
Neural Computing and Applications Aims and scope Submit manuscript

Abstract

Nearly one in five families in the country was poor in 2012, according to the Philippine Statistics Authority. While this proportion is lower than the corresponding figures from 2006 and 2009, the absolute number of poor families has actually grown from 3.8 million in 2006 to 4.2 million in 2012 due to the increase in population. Using data samples that have been collected from 69,130 households through a comprehensive community-based monitoring survey conducted in one of the cities that comprise Metro Manila, we attempt to identify the characteristics that differentiate between poor and non-poor households. Using back-propagation neural networks, we are able to correctly predict 73% of the poor households and 60% of the non-poor households. Moreover, the rules extracted from one of these networks provide concise description of how households are classified as poor based on their demographic characteristics and information pertaining to their surrounding living conditions.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1

Similar content being viewed by others

References

  1. Achia TBO, Wangombe A, Khadioli N (2010) A logistic regression model to identify key determinants of poverty using demographic and health survey data. Eur J Soc Sci 13(1):38–45

    Google Scholar 

  2. Adeli H (2001) Neural networks in civil engineering. Comput-Aided Civil Infrastruct Eng 16(2):126–142

    Article  Google Scholar 

  3. Alem Y, Köhlin G, Stage J (2013) The persistence of subjective poverty in urban Ethiopia. World Dev 56:51–61

    Article  Google Scholar 

  4. Ansoms A, McKay A (2010) A quantitative analysis of poverty and livelihood profiles: the case of rural Rwanda. Food Policy 35:584–598

    Article  Google Scholar 

  5. Babu SC, Sanyal P (2009) Food security, poverty and nutrition policy analysis: statistical methods and applications. Academic Press, Burlington

    Google Scholar 

  6. Belhadj B (2011) New fuzzy indices of poverty by distinguishing three levels of poverty. Res Econ 65:221–231

    Article  Google Scholar 

  7. Benson T, Chamberlin J, Rhinehart I (2005) An investigation of the spatial determinants of the local prevalence of poverty in rural Malawi. Food Policy 30:532–550

    Article  Google Scholar 

  8. Bigman D, Srinivasan PV (2002) Geographical targeting of poverty alleviation programs: methodology and applications in rural India. J Policy Model 24:237–255

    Article  Google Scholar 

  9. Chapoto A, Kirimi L, Kadilaya S (2012) Poverty and prime-age mortality in eastern and southern Africa: evidence from Zambia and Kenya. World Dev 40(9):1839–1853

    Article  Google Scholar 

  10. Dennis JE Jr, Schnabel RE (1983) Numerical methods for unconstrained optimization and nonlinear equations. Prentice Halls, Englewood Cliffs

    MATH  Google Scholar 

  11. Duy VQ, D’Haese M, Lemba J, Hau LL, D’Haese L (2011) Determinants of household access to formal credit in the rural areas of the Mekong delta, Vietnam. Afr Asian Stud 11:261–287

    Article  Google Scholar 

  12. Erguzel TT, Sayar GH, Tarhan N (2016) Artificial intelligence approach to classify unipolar and bipolar depressive disorders. Neural Comput Appl 27(6):1607–1616

    Article  Google Scholar 

  13. Fernández-Gámez MA, Gil-Corral AM, Galán-Valdivieso F (2016) Corporate reputation and market value: evidence with generalized regression neural networks. Expert Syst Appl 46:69–76

    Article  Google Scholar 

  14. Gounder R, Xing Z (2012) Impact of education and health on poverty reduction: monetary and non-monetary evidence from Fiji. Econ Model 29:787–794

    Article  Google Scholar 

  15. Haykin S (1999) Neural networks: a comprehensive foundation, 2nd edn. Prentice Hall, Englewood Cliffs

    MATH  Google Scholar 

  16. Hoogeveen JG (2005) Measuring welfare for small but vulnerable groups: poverty and disability in Uganda. J Afr Econ 14(4):603–631

    Article  Google Scholar 

  17. Huang W, Lai KK, Nakamori Y, Wang S (2004) Forecasting foreign exchange rates with artificial neural networks: a review. Int J Inf Technol Decis Mak 03(1):145–165

    Article  Google Scholar 

  18. Jha GK, Kanchan S (2014) Time-delay neural networks for time series prediction: an application to the monthly wholesale price of oilseeds in India. Neural Comput Appl 24(3):563–571

    Article  Google Scholar 

  19. Kalogirou SA (2001) Artificial neural networks in renewable energy systems applications: a review. Renew Sustain Energy Rev 5(4):373–401

    Article  Google Scholar 

  20. Kiakojoori S, Khorasani K (2016) Dynamic neural networks for gas turbine engine degradation, health monitoring and prognosis. Neural Comput Appl 27(8):2157–2192

    Article  Google Scholar 

  21. Lin S-M (2013) Analysis of service satisfaction in web auction logistics service using a combination of Fruit fly optimization algorithm and general regression neural network. Neural Comput Appl 22(3):783–791

    Article  Google Scholar 

  22. Lisboa PJG (2002) A review of evidence of health benefit from artificial neural networks in medical intervention. Neural Netw 15(1):11–39

    Article  Google Scholar 

  23. Lisboa PJ, Taktak AFG (2006) The use of artificial neural networks in decision support in cancer: a systematic review. Neural Netw 19(4):408–415

    Article  Google Scholar 

  24. Liu H, Tan ST (1995) X2R: a fast rule generator. In: Proceedings of IEEE international conference on systems, man and cybernetics, pp 1631–1635

  25. Maharjan KL, Joshi NP (2011) Determinants of household food security in Nepal: a binary logistic regression analysis. J Mt Sci 8:403–413

    Article  Google Scholar 

  26. Maier HG, Dandy GC (2000) Neural networks for the prediction and forecasting of water resources variables: a review of modelling issues and applications. Environ Model Softw 15(1):101–124

    Article  Google Scholar 

  27. Mitra S (2013) Early warning prediction system for high inflation: an elitist neuro-genetic network model for the Indian economy. Neural Comput Appl 22(S1):447–462

    Article  Google Scholar 

  28. Neff D (2013) Fuzzy set theoretic applications in poverty research. Policy Soc 32:319–331

    Article  Google Scholar 

  29. Paliwal M, Kumar UA (2009) Neural networks and statistical techniques: a review of applications. Expert Syst Appl 36:2–17

    Article  Google Scholar 

  30. Pany PK, Ghoshal SP (2015) Dynamic electricity price forecasting using local linear wavelet neural network. Neural Comput Appl 26(8):2039–2047

    Article  Google Scholar 

  31. Ravallion M (1998) Poverty lines in theory and practice. Living Standard measurement study working paper no. 133. World Bank, Washington DC

  32. Senf C, Lakes T (2012) Comparing support vector regression and statistical linear regression for predicting poverty incidence in Vietnam. In: Bridging the geographic information sciences, lecture notes in geoinformation and cartography, pp 251–265

    Google Scholar 

  33. Setiono R (1997) A penalty-function approach for pruning feedforward neural networks. Neural Comput 9(1):185–204

    Article  Google Scholar 

  34. Setiono R, Liu H (1996) Symbolic representation of neural networks. IEEE Comput 29(3):71–77

    Article  Google Scholar 

  35. Sy I (2013) The subjective approach as a tool for understanding: The case of Senegal. Proc Econ Financ 5:336–345

    Article  Google Scholar 

  36. Vellido V, Lisboa PJG, Vaughn J (1999) Neural networks in business: a survey of applications. Expert Syst Appl 17(1):51–70

    Article  Google Scholar 

  37. Woolard I, Klasen S (2005) Determinants of income mobility and poverty dynamics in South Africa. J Dev Stud 41(5):865–897

    Article  Google Scholar 

  38. Yaseen ZM, El-Shafie A, Afan HA, Hameed M (2016) RBFNN versus FFNN for daily river flow forecasting at Johor River, Malaysia. Neural Comput Appl 27(6):1533–1542

    Article  Google Scholar 

  39. Yazici B, Memmedli M, Aslanargun A, Sama S (2010) Analysis of international debt problem using artificial neural networks and statistical methods. Neural Comput Appl 19(8):1207–1216

    Article  Google Scholar 

  40. Zhang Z, Friedrich K (2003) Artificial neural networks applications to polymer composites: a review. Compos Sci Technol 63(14):2029–2044

    Article  Google Scholar 

Download references

Acknowledgements

The authors wish to thank the Angelo King Institute (AKI) of De La Salle University, under whom the CBMS project is lodged, for sharing the raw survey data from Pasay City.

Author information

Authors and Affiliations

  1. College of Computer Studies, De La Salle University, 2401 Taft Avenue, Manila, Philippines

    Arnulfo Azcarraga

  2. School of Computing, National University of Singapore, 13 Computing Drive, Singapore, 117417, Republic of Singapore

    Rudy Setiono

Authors
  1. Arnulfo Azcarraga
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Rudy Setiono
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Rudy Setiono.

Ethics declarations

Conflicts of interest

The authors declare that they have no conflict of interest.

Appendix

Appendix

List of variables included in the study to distinguish between poor and non-poor households.

Group 1: Demographic information of the head of the household

  1. 1.

    Gender: 1 \(=\) male, 0 \(=\) otherwise.

  2. 2.

    Age: continuous.

  3. 3.

    Civil status: single, legally married, widowed, separated, common law marriage, unknown.

  4. 4.

    Religion: 1 \(=\) Roman Catholic, 2 \(=\) Protestant, 3 \(=\) local non-Catholic church, 4 \(=\) local Catholic-rite church, 5 \(=\) Islam, 6 \(=\) others, 7 \(=\) no religion, 8 \(=\) unknown.

  5. 5.

    Education: 1 \(=\) elementary school, 2 \(=\) secondary school, 3 \(=\) post-secondary, 4 \(=\) college graduate, 5 \(=\) post-graduate.

  6. 6.

    Literacy: 1 \(=\) literate, 0 \(=\) otherwise.

  7. 7.

    Job: 1 \(=\) currently employed, 0 \(=\) otherwise.

  8. 8.

    Phil Health: 1 \(=\) member of Philippine Health, 0 \(=\) otherwise.

Group 2: Other information about the household

  1. 1.

    Foreign worker in family: 1 \(=\) at least one member of the family is currently working overseas, 0 \(=\) otherwise.

  2. 2.

    Single parent: 1 \(=\) single parent, \(=\) 0 otherwise.

  3. 3.

    Board passer: 1 \(=\) if there is a board exam passer in the family (e.g., nurse, engineering, architecture), 0 \(=\) otherwise.

  4. 4.

    Disable: 1 \(=\) if there is a disable family member, 0 \(=\) otherwise.

  5. 5.

    Death-indicator: 1 \(=\) if there was a death in the family in the past year, 0 \(=\) otherwise.

  6. 6.

    Calamity-indicator: 1 \(=\) if the family experienced a calamity (e.g., typhoon) in the past year, 0 \(=\) otherwise.

  7. 7.

    FoodShortage-indicator: 1 \(=\) the family experienced hunger in the past year, 0 \(=\) otherwise.

  8. 8.

    Garbage: 1 \(=\) if there is an organized (city, town, subdivision) garbage collection, 0 \(=\) otherwise.

Group 3: Access to water source (binary variable)

  1. 1.

    Community water system—own use

  2. 2.

    Community water system—shared with other households

  3. 3.

    Artesian deep well—own use

  4. 4.

    Artesian deep well—shared with other households

  5. 5.

    Artesian shallow well—own use

  6. 6.

    Artesian shallow well—shared with other households

  7. 7.

    Dug/shallow well—own use

  8. 8.

    Dug/shallow well—shared with other households

  9. 9.

    River, stream, lake, spring and other bodies of water

  10. 10.

    Bottled water/purified/distilled water

  11. 11.

    Tanker truck/peddler

  12. 12.

    Others

Group 4: Access to water distribution (binary variable)

  1. 1.

    Unknown

  2. 2.

    Within premises

  3. 3.

    Outside premises but 250 m or less

  4. 4.

    Outside premises 251 m or more

  5. 5.

    don’t know

Group 5: Access to toilet (binary variable)

  1. 1.

    Water sealed flush to sewerage system/septic tank—own use

  2. 2.

    Water sealed flush to sewerage system/septic tank—shared with other households

  3. 3.

    Closed pit

  4. 4.

    Open pit

  5. 5.

    No toilet

  6. 6.

    Others

Group 6: Dwelling wall (binary variable)

  1. 1.

    Strong materials

  2. 2.

    Light materials

  3. 3.

    Salvaged/makeshift materials

  4. 4.

    Mixed but predominantly strong materials

  5. 5.

    Mixed but predominantly light materials

  6. 6.

    Mixed but predominantly salvaged materials

Group 7: Dwelling roof (binary variable)

  1. 1.

    Strong materials

  2. 2.

    Light materials

  3. 3.

    Salvaged/makeshift materials

  4. 4.

    Mixed but predominantly strong materials

  5. 5.

    Mixed but predominantly light materials

  6. 6.

    Mixed but predominantly salvaged materials

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Azcarraga, A., Setiono, R. Neural network rule extraction for gaining insight into the characteristics of poverty. Neural Comput & Applic 30, 2795–2806 (2018). https://doi.org/10.1007/s00521-017-2889-8

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00521-017-2889-8

Keywords

Search

Navigation