Top

BMC Medical Informatics and Decision Making

Published in:

Open Access 01-12-2023 | Lyme Disease | Research

Leveraging machine learning approaches for predicting potential Lyme disease cases and incidence rates in the United States using Twitter

Authors: Srikanth Boligarla, Elda Kokoè Elolo Laison, Jiaxin Li, Raja Mahadevan, Austen Ng, Yangming Lin, Mamadou Yamar Thioub, Bruce Huang, Mohamed Hamza Ibrahim, Bouchra Nasri

Published in: BMC Medical Informatics and Decision Making | Issue 1/2023

Abstract

Background

Lyme disease is one of the most commonly reported infectious diseases in the United States (US), accounting for more than \(90\%\) of all vector-borne diseases in North America.

Objective

In this paper, self-reported tweets on Twitter were analyzed in order to predict potential Lyme disease cases and accurately assess incidence rates in the US.

Methods

The study was done in three stages: (1) Approximately 1.3 million tweets were collected and pre-processed to extract the most relevant Lyme disease tweets with geolocations. A subset of tweets were semi-automatically labelled as relevant or irrelevant to Lyme disease using a set of precise keywords, and the remaining portion were manually labelled, yielding a curated labelled dataset of 77, 500 tweets. (2) This labelled data set was used to train, validate, and test various combinations of NLP word embedding methods and prominent ML classification models, such as TF-IDF and logistic regression, Word2vec and XGboost, and BERTweet, among others, to identify potential Lyme disease tweets. (3) Lastly, the presence of spatio-temporal patterns in the US over a 10-year period were studied.

Results

Preliminary results showed that BERTweet outperformed all tested NLP classifiers for identifying Lyme disease tweets, achieving the highest classification accuracy and F1-score of \(90\%\). There was also a consistent pattern indicating that the West and Northeast regions of the US had a higher tweet rate over time.

Conclusions

We focused on the less-studied problem of using Twitter data as a surveillance tool for Lyme disease in the US. Several crucial findings have emerged from the study. First, there is a fairly strong correlation between classified tweet counts and Lyme disease counts, with both following similar trends. Second, in 2015 and early 2016, the social media network like Twitter was essential in raising popular awareness of Lyme disease. Third, counties with a high incidence rate were not necessarily related with a high tweet rate, and vice versa. Fourth, BERTweet can be used as a reliable NLP classifier for detecting relevant Lyme disease tweets.

Available only for authorised users

Murphree Bacon R, Kugeler KJ, Mead PS. Surveillance for Lyme disease--United States, 1992-2006. 2008.

Kugeler KJ, Schwartz AM, Delorey MJ, Mead PS, Hinckley AF. Estimating the Frequency of Lyme Disease Diagnoses, United States, 2010–2018. Emerg Infect Dis. 2021;27(2):616–9. https://doi.org/10.3201/eid2702.202731. Accessed 17 Sep 2022.

Kumar D, Downs LP, Adegoke A, Machtinger E, Oggenfuss K, Ostfeld RS, et al. An Exploratory Study on the Microbiome of Northern and Southern Populations of Ixodes scapularis Ticks Predicts Changes and Unique Bacterial Interactions. Pathogens. 2022;11(2):130. https://doi.org/10.3390/pathogens11020130. Accessed 17 Sep 2022.

Marques AR, Strle F, Wormser GP. Comparison of Lyme Disease in the United States and Europe. Emerg Infect Dis. 2021;27(8):2017–2024. https://doi.org/10.3201/eid2708.204763. Accessed 17 Sep 2022.

Davidsson M. The Financial Implications of a Well-Hidden and Ignored Chronic Lyme Disease Pandemic. Healthcare. 2018;6(1):16. https://doi.org/10.3390/healthcare6010016. Accessed 17 Sep 2022.

Hook SA, Jeon S, Niesobecki SA, Hansen AP, Meek JI, Bjork JKH, et al. Economic Burden of Reported Lyme Disease in High-Incidence Areas, United States, 2014–2016. Emerg Infect Dis. 2022;28(6). https://doi.org/10.3201/eid2806.211335. Accessed 17 Sep 2022.

Mead PS. Epidemiology of Lyme Disease. Infect Dis Clin N Am. 2015;29(2):187–210. https://doi.org/10.1016/j.idc.2015.02.010. Accessed 17 Sep 2022.

Stanek G, Wormser GP, Gray J, Strle F. Lyme borreliosis. Lancet. 2012;379(9814):461–473. https://doi.org/10.1016/S0140-6736(11)60103-7. Accessed 17 Sep 2022.

Piesman J, Gern L. Lyme borreliosis in Europe and North America. Parasitology. 2004;129(S1):S191–S220. https://doi.org/10.1017/S0031182003004694. Accessed 17 Sep 2022.

10.

Rochlin I, Ninivaggi DV, Benach JL. Malaria and Lyme disease - the largest vector-borne US epidemics in the last 100 years: success and failure of public health. BMC Public Health. 2019;19(1):804. https://doi.org/10.1186/s12889-019-7069-6. Accessed 17 Sep 2022.

11.

Aenishaenslin C, Bouchard C, Koffi JK, Pelcat Y, Ogden NH. Evidence of rapid changes in Lyme disease awareness in Canada. Ticks Tick-Borne Dis. 2016;7(6):1067–1074. https://doi.org/10.1016/j.ttbdis.2016.09.007. Accessed 17 Sep 2022.

12.

Alkishe A, Raghavan RK, Peterson AT. Likely Geographic Distributional Shifts among Medically Important Tick Species and Tick-Associated Diseases under Climate Change in North America: A Review. Insects. 2021;12(3):225. https://doi.org/10.3390/insects12030225. Accessed 17 Sep 2022.

13.

Ogden NH, Feil EJ, Leighton PA, Lindsay LR, Margos G, Mechai S, et al. Evolutionary Aspects of Emerging Lyme Disease in Canada. Appl Environ Microbiol. 2015;81(21):7350–9. https://doi.org/10.1128/AEM.01671-15.CrossRefPubMedCentralPubMed

14.

Brinkerhoff R, Kitron U, Diuk-Wasser MA, Fish D, Melton F, Cislo P, et al. Human Risk of Infection with Borrelia burgdorferi, the Lyme Disease Agent, in Eastern United States. Am J Trop Med Hyg. 2012;86(2):320–327. https://doi.org/10.4269/ajtmh.2012.11-0395. Accessed 17 Sep 2022.

15.

Kilpatrick AM, Dobson ADM, Levi T, Salkeld DJ, Swei A, Ginsberg HS, et al. Lyme disease ecology in a changing world: consensus, uncertainty and critical gaps for improving control. Phil Trans R Soc B Biol Sci. 2017;372(1722):20160117. https://doi.org/10.1098/rstb.2016.0117. Accessed 17 Sep 2022.

16.

Kugeler KJ, Eisen RJ. Challenges in Predicting Lyme Disease Risk. JAMA Netw Open. 2020;3(3): e200328. https://doi.org/10.1001/jamanetworkopen.2020.0328.CrossRefPubMedCentralPubMed

17.

Eisen RJ, Eisen L, Ogden NH, Beard CB. Linkages of Weather and Climate With Ixodes scapularis and Ixodes pacificus (Acari: Ixodidae), Enzootic Transmission of Borrelia burgdorferi, and Lyme Disease in North America. J Med Entomol. 2016;53(2):250–61. https://doi.org/10.1093/jme/tjv199.CrossRefPubMed

18.

Wood CL, Lafferty KD. Biodiversity and disease: a synthesis of ecological perspectives on Lyme disease transmission. Trends Ecol Evol. 2013;28(4):239–247. https://doi.org/10.1016/j.tree.2012.10.011. Accessed 17 Sep 2022.

19.

Eisen RJ, Piesman J, Zielinski-Gutierrez E, Eisen L. What Do We Need to Know About Disease Ecology to Prevent Lyme Disease in the Northeastern United States?: Table 1. J Med Entomol. 2012;49(1):11–22. https://doi.org/10.1603/ME11138. Accessed 17 Sep 2022.

20.

Eisen RJ, Eisen L. The Blacklegged Tick, Ixodes scapularis : An Increasing Public Health Concern. Trends Parasitol. 2018;34(4):295–309. https://doi.org/10.1016/j.pt.2017.12.006. Accessed 17 Sep 2022.

21.

Bouchard C, Beauchamp G, Leighton PA, Lindsay R, Bélanger D, Ogden NH. Does high biodiversity reduce the risk of Lyme disease invasion? Parasites Vectors. 2013;6(1):195. https://doi.org/10.1186/1756-3305-6-195. Accessed 17 Sep 2022.

22.

Brownstein JS, Skelly DK, Holford TR, Fish D. Forest fragmentation predicts local scale heterogeneity of Lyme disease risk. Oecologia. 2005;146(3):469–475. https://doi.org/10.1007/s00442-005-0251-9. Accessed 17 Sep 2022.

23.

Lantos PM, Tsao J, Janko M, Arab A, von Fricken ME, Auwaerter PG, et al. Environmental Correlates of Lyme Disease Emergence in Southwest Virginia, 2005–2014. J Med Entomol. 2021;58(4):1680–1685. https://doi.org/10.1093/jme/tjab038. Accessed 17 Sep 2022.

24.

Steere AC. Lyme Disease. N Engl J Med. 2001;345(2):115–125. https://doi.org/10.1056/NEJM200107123450207. Accessed 17 Sep 2022.

25.

Borchers AT, Keen CL, Huntley AC, Gershwin ME. Lyme disease: A rigorous review of diagnostic criteria and treatment. J Autoimmun. 2015;57:82–115. https://doi.org/10.1016/j.jaut.2014.09.004. Accessed 17 Sep 2022.

26.

Carriveau A, Poole H, Thomas A. Lyme Disease. Nurs Clin N Am. 2019;54(2):261–75. https://doi.org/10.1016/j.cnur.2019.02.003. Accessed 17 Sep 2022.

27.

Chomel B. Lyme disease: -EN- -FR- La maladie de Lyme -ES- Enfermedad de Lyme. Rev Sci Tech l’OIE. 2015 34(2):569–76. https://doi.org/10.20506/rst.34.2.2380. Accessed 17 Sep 2022.

28.

Rebman AW, Yang T, Mihm EA, Novak CB, Yoon I, Powell D, et al. The presenting characteristics of erythema migrans vary by age, sex, duration, and body location. Infection. 2021;49(4):685–92. https://doi.org/10.1007/s15010-021-01590-0. Accessed 17 Sep 2022.

29.

Bockenstedt LK, Wormser GP. Review: Unraveling Lyme Disease: Lyme Disease. Arthritis Rheumatol. 2014;66(9):2313–23. https://doi.org/10.1002/art.38756. Accessed 17 Sep 2022.

30.

Ross Russell AL, Dryden MS, Pinto AA, Lovett JK. Lyme disease: diagnosis and management. Pract Neurol. 2018;18(6):455–64. https://doi.org/10.1136/practneurol-2018-001998. Accessed 17 Sep 2022.

31.

Sanchez JL. Clinical Manifestations and Treatment of Lyme Disease. Clin Lab Med. 2015;35(4):765–78. https://doi.org/10.1016/j.cll.2015.08.004. Accessed 17 Sep 2022.

32.

Schoen RT. Challenges in the Diagnosis and Treatment of Lyme Disease. Curr Rheumatol Rep. 2020;22(1):3. https://doi.org/10.1007/s11926-019-0857-2. Accessed 17 Sep 2022.

33.

Schoen RT. Lyme disease: diagnosis and treatment. Curr Opin Rheumatol. 2020;32(3):247–54. https://doi.org/10.1097/BOR.0000000000000698. Accessed 17 Sep 2022.

34.

Aguero-Rosenfeld ME, Wang G, Schwartz I, Wormser GP. Diagnosis of Lyme Borreliosis. Clin Microbiol Rev. 2005;18(3):484–509. https://doi.org/10.1128/CMR.18.3.484-509.2005. Accessed 17 Sep 2022.

35.

Clark RP, Hu LT. Prevention of Lyme Disease and Other Tick-Borne Infections. Infect Dis Clin N Am. 2008;22(3):381–96. https://doi.org/10.1016/j.idc.2008.03.007. Accessed 17 Sep 2022.

36.

Raizman EA, Holland JD, Shukle JT. White-Tailed Deer ( Odocoileus virginianus ) as a Potential Sentinel for Human Lyme Disease in Indiana: Deer as a Sentinel for Human Lyme Disease. Zoonoses Public Health. 2013;60(3):227–33. https://doi.org/10.1111/j.1863-2378.2012.01518.x. Accessed 17 Sep 2022.

37.

Campbell GL, Fritz CL, Fish D, Nowakowski J, Nadelman RB, Wormser GP. Estimation of the Incidence of Lyme Disease. Am J Epidemiol. 1998;148(10):1018–26. https://doi.org/10.1093/oxfordjournals.aje.a009568. Accessed 17 Sep 2022.

38.

Coyle BS, Strickland GT, Liang YY, Pena C, McCarter R, Israel E. The Public Health Impact of Lyme Disease in Maryland. J Infect Dis. 1996;173(5):1260–2. https://doi.org/10.1093/infdis/173.5.1260. Accessed 17 Sep 2022.

39.

Ertel SH, Nelson RS, Cartter ML. Effect of Surveillance Method on Reported Characteristics of Lyme Disease, Connecticut, 1996–2007. Emerg Infect Dis. 2012;18(2):242–7. https://doi.org/10.3201/eid1802.101219. Accessed 17 Sep 2022.

40.

Lou Y, Wu J. Modeling Lyme disease transmission. Infect Dis Model. 2017;2(2):229–43. https://doi.org/10.1016/j.idm.2017.05.002. Accessed 17 Sep 2022.

41.

Meek JI, Roberts CL, Smith EV, Cartter ML. Underreporting of Lyme Disease by Connecticut Physicians, 1992. J Public Health Manag Pract. 1996;2(4):61–5. https://doi.org/10.1097/00124784-199623000-00017. Accessed 17 Sep 2022.

42.

Naleway AL. Lyme Disease Incidence in Wisconsin: A Comparison of State-reported Rates and Rates from a Population-based Cohort. Am J Epidemiol. 2002;155(12):1120–7. https://doi.org/10.1093/aje/155.12.1120. Accessed 17 Sep 2022.

43.

Nelson CA, Saha S, Kugeler KJ, Delorey MJ, Shankar MB, Hinckley AF, et al. Incidence of Clinician-Diagnosed Lyme Disease, United States, 2005–2010. Emerg Infect Dis. 2015;21(9):1625–31. https://doi.org/10.3201/eid2109.150417. Accessed 17 Sep 2022.

44.

Ratti V, Winter JM, Wallace DI. Dilution and amplification effects in Lyme disease: Modeling the effects of reservoir-incompetent hosts on Borrelia burgdorferi sensu stricto transmission. Ticks Tick-Borne Dis. 2021;12(4):101724. https://doi.org/10.1016/j.ttbdis.2021.101724. Accessed 17 Sep 2022.

45.

Schwartz AM, Kugeler KJ, Nelson CA, Marx GE, Hinckley AF. Use of Commercial Claims Data for Evaluating Trends in Lyme Disease Diagnoses, United States, 2010–2018. Emerg Infect Dis. 2021;27(2):499–507. https://doi.org/10.3201/eid2702.202728. Accessed 17 Sep 2022.

46.

Stevens LK, Kolivras KN, Hong Y, Thomas VA, Campbell JB, Prisley SP. Future Lyme disease risk in the south-eastern United States based on projected land cover. Geospatial Health. 2019;14(1). https://doi.org/10.4081/gh.2019.751. Accessed 17 Sep 2022.

47.

Kobayashi T, Higgins Y, Melia MT, Auwaerter PG. Mistaken Identity: Many Diagnoses are Frequently Misattributed to Lyme Disease. Am J Med. 2022;135(4):503–5115. https://doi.org/10.1016/j.amjmed.2021.10.040. Accessed 17 Sep 2022.

48.

Eysenbach G. Infodemiology and Infoveillance: Framework for an Emerging Set of Public Health Informatics Methods to Analyze Search, Communication and Publication Behavior on the Internet. J Med Internet Res. 2009;11(1):11. https://doi.org/10.2196/jmir.1157. Accessed 17 Sep 2022.

49.

Pereira-Sanchez V, Alvarez-Mon MA, Del Barco AA, Alvarez-Mon M, Teo A, et al. Exploring the extent of the hikikomori phenomenon on twitter: Mixed methods study of western language tweets. J Med Internet Res. 2019;21(5):14167.CrossRef

50.

Thiebaut R, Thiessard F, et al. Public health and epidemiology informatics. Yearb Med Inform. 2017;26(01):248–51.CrossRefPubMedCentralPubMed

51.

Allen C, Tsou MH, Aslam A, Nagel A, Gawron JM. Applying GIS and Machine Learning Methods to Twitter Data for Multiscale Surveillance of Influenza. PLoS ONE. 2016;11(7):0157734. https://doi.org/10.1371/journal.pone.0157734. Accessed 17 Sep 2022.

52.

Mavragani A, Ochoa G, et al. Google Trends in infodemiology and infoveillance: methodology framework. JMIR Public Health Surveill. 2019;5(2):13439.CrossRef

53.

Nuti SV, Wayda B, Ranasinghe I, Wang S, Dreyer RP, Chen SI, et al. The use of google trends in health care research: a systematic review. PLoS ONE. 2014;9(10):109583.CrossRef

54.

Bowman LJ. Statista. J Bus Finance Librariansh. 2022;27(4):304–9. https://doi.org/10.1080/08963568.2022.2087018.

55.

Arias M, Arratia A, Xuriguera R. Forecasting with twitter data. ACM Trans Intell Syst Technol. 2013;5(1):1–24. https://doi.org/10.1145/2542182.2542190. Accessed 17 Sep 2022.

56.

Aslam AA, Tsou MH, Spitzberg BH, An L, Gawron JM, Gupta DK, et al. The Reliability of Tweets as a Supplementary Method of Seasonal Influenza Surveillance. J Med Internet Res. 2014;16(11):250. https://doi.org/10.2196/jmir.3532. Accessed 17 Sep 2022.

57.

Kramer CM, Barkhausen J, Flamm SD, Kim RJ, Nagel E. Standardized cardiovascular magnetic resonance (CMR) protocols 2013 update. J Cardiovasc Magn Reson. 2013;15(1):1–10.CrossRef

58.

Pollett S, Althouse BM, Forshey B, Rutherford GW, Jarman RG. Internet-based biosurveillance methods for vector-borne diseases: Are they novel public health tools or just novelties? PLoS Negl Trop Dis. 2017;11(11):0005871.CrossRef

59.

Basch CH, Mullican LA, Boone KD, Yin J, Berdnik A, Eremeeva ME, et al. Lyme Disease and YouTubeTM: A Cross-Sectional Study of Video Contents. Osong Public Health Res Perspect. 2017;8(4):289–92. https://doi.org/10.24171/j.phrp.2017.8.4.10. Accessed 17 Sep 2022.

60.

Kapitány-Fövény M, Ferenci T, Sulyok Z, Kegele J, Richter H, Vályi-Nagy I, et al. Can Google Trends data improve forecasting of Lyme disease incidence? Zoonoses Public Health. 2019;66(1):101–7. https://doi.org/10.1111/zph.12539. Accessed 17 Sep 2022.

61.

Kim D, Maxwell S, Le Q. Spatial and Temporal Comparison of Perceived Risks and Confirmed Cases of Lyme Disease: An Exploratory Study of Google Trends. Front Public Health. 2020;8:395. https://doi.org/10.3389/fpubh.2020.00395. Accessed 17 Sep 2022.

62.

Kutera M, Berke O, Sobkowich K. Spatial epidemiological analysis of Lyme disease in southern Ontario utilizing Google Trends searches. Environ Health Rev. 2021;64(4):105–10. https://doi.org/10.5864/d2021-025. Accessed 17 Sep 2022.

63.

Pesälä S, Virtanen MJ, Sane J, Mustonen P, Kaila M, Helve O. Health Information–Seeking Patterns of the General Public and Indications for Disease Surveillance: Register-Based Study Using Lyme Disease. JMIR Public Health Surveill. 2017;3(4):86. https://doi.org/10.2196/publichealth.8306. Accessed 17 Sep 2022.

64.

Sadilek A, Hswen Y, Bavadekar S, Shekel T, Brownstein JS, Gabrilovich E. Lymelight: forecasting Lyme disease risk using web search data. NPJ Digit Med. 2020;3(1):16. https://doi.org/10.1038/s41746-020-0222-x. Accessed 17 Sep 2022.

65.

Scheerer C, Rüth M, Tizek L, Köberle M, Biedermann T, Zink A. Googling for Ticks and Borreliosis in Germany: Nationwide Google Search Analysis From 2015 to 2018. J Med Internet Res. 2020;22(10):18581. https://doi.org/10.2196/18581. Accessed 17 Sep 2022.

66.

Seifter A, Schwarzwalder A, Geis K, Aucott J. The utility of “Google Trends” for epidemiological research: Lyme disease as an example. Geospatial Health. 2010;4(2):135. https://doi.org/10.4081/gh.2010.195. Accessed 17 Sep 2022.

67.

Sulyok M, Richter H, Sulyok Z, Kapitány-Fövény M, Walker MD. Predicting tick-borne encephalitis using Google Trends. Ticks Tick-Borne Dis. 2020;11(1):101306. https://doi.org/10.1016/j.ttbdis.2019.101306. Accessed 17 Sep 2022.

68.

Tulloch JSP, Vivancos R, Christley RM, Radford AD, Warner JC. Mapping tweets to a known disease epidemiology; a case study of Lyme disease in the United Kingdom and Republic of Ireland. J Biomed Inform. 2019;100:100060. https://doi.org/10.1016/j.yjbinx.2019.100060. Accessed 17 Sep 2022.

69.

Yiannakoulias N, Tooby R, Sturrock SL. Celebrity over science? An analysis of Lyme disease video content on YouTube. Soc Sci Med. 2017;191:57–60. https://doi.org/10.1016/j.socscimed.2017.08.042. Accessed 17 Sep 2022.

70.

Mikolov T, Chen K, Corrado G, Dean J. Efficient estimation of word representations in vector space. arXiv preprint arXiv:1301.3781. 2013.

71.

Chen T, Guestrin C. Xgboost: A scalable tree boosting system. In Proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining; 2016. pp. 785-794.

72.

Rehurek R, Sojka P. Gensim–python framework for vector space modelling. NLP Centre, Faculty of Informatics, Masaryk University, Brno, Czech Republic. 2011;3(2):2.

73.

Ramos J, et al. Using tf-idf to determine word relevance in document queries. In: Proceedings of the first instructional conference on machine learning, vol 242. Citeseer; 2003. p. 29–48.

74.

Menard S. Applied logistic regression analysis, vol 106. Sage; 2002.

75.

Devlin J, Chang MW, Lee K, Toutanova K. Bert: Pre-training of deep bidirectional transformers for language understanding. arXiv preprint arXiv:1810.04805. 2018.

76.

Nguyen DQ, Vu T, Nguyen AT. BERTweet: A pre-trained language model for English Tweets. arXiv preprint arXiv:2005.10200. 2020.

Title: Leveraging machine learning approaches for predicting potential Lyme disease cases and incidence rates in the United States using Twitter
Authors: Srikanth Boligarla
Elda Kokoè Elolo Laison
Jiaxin Li
Raja Mahadevan
Austen Ng
Yangming Lin
Mamadou Yamar Thioub
Bruce Huang
Mohamed Hamza Ibrahim
Bouchra Nasri
Publication date: 01-12-2023
Publisher: BioMed Central
Keywords: Lyme Disease
Tick
Tick
Published in: BMC Medical Informatics and Decision Making / Issue 1/2023
Electronic ISSN: 1472-6947
DOI: https://doi.org/10.1186/s12911-023-02315-z

Keynote webinar | Spotlight on medication adherence

Springer Medicine

Leveraging machine learning approaches for predicting potential Lyme disease cases and incidence rates in the United States using Twitter

Abstract

Background

Objective

Methods

Results

Conclusions

Keynote webinar | Spotlight on medication adherence

Springer Medicine

Abstract

Background

Objective

Methods

Results

Conclusions

Please log in to get access to this content

Other articles of this Issue 1/2023

Digital health literacy and digital engagement for people with severe mental ill health across the course of the COVID-19 pandemic in England

Antecedents predicting digital contact tracing acceptance: a systematic review and meta-analysis

Determining the content and needs assessment a mobile-based self-care program in infertile men

Reducing prescribing of antibiotics for acute respiratory infections using a frontline nurse-led EHR-Integrated clinical decision support tool: protocol for a stepped wedge randomized control trial

Comparison of causal forest and regression-based approaches to evaluate treatment effect heterogeneity: an application for type 2 diabetes precision medicine

Predicting the drop out from the maternal, newborn and child healthcare continuum in three East African Community countries: application of machine learning models