Realizing Social-Media-Based Analytics for Smart Agriculture
- 14 Downloads
Continuous fluctuation of the daily prices of vegetables and pulses is destabilizing to a country’s economy and hinders the nation’s monetary planning. One factor that affects the fluctuation of the prices is changes in the production of the vegetables and pulses. Early detection and treatment of crop and plant diseases will substantially improve production and constitutes an attempt to introduce smart technology into the system. There is an immediate need for an intelligent solution that advises farmers on how to address crop and plant diseases at the point of demand. In this paper, we have explored social media analytics as a knowledge-intensive process and proposed a new framework for identifying the relevant details of plants from both text and images. This approach will generate knowledge about plant diseases and other areas of development that are related to precision agriculture. The community of farmers within social media is expected to provide immediate suggestions to questions that are posted by members regarding plant diseases and other problems. The newly designed innovative framework identifies the relevant features, finds similarities among the farmers’ heuristics, and ranks their suggestions related to merit and relevance. Our system employs deep learning, natural language processing and other predictive models to compare the combination of text and image extracts and suggest a suitable solution to the farmer’s query. The efficiency of the employed models is evaluated to avoid false positives, and the models are implemented on social media to address plant diseases and other relevant details as a part of evolving smart agriculture.
KeywordsSocial media Plant disease monitoring Deep learning Rank regression Smart agriculture
- 2.World Bank. (2018) Overview, [online]. Available at http://www.worldbank.org/en/topic/agriculture/overview Accessed 26 Mar 2018.
- 3.A. Brian. (1987). Demand forecasting and estimation. 77–85.Google Scholar
- 5.Fan, T. (2013). Smart agriculture based on cloud computing and IOT. Journal of Convergence Information Technology., 8, 2.Google Scholar
- 13.Jeff Gentry (2015). twitteR: R Based Twitter Client. R, package version 1.1.9, https://CRAN.R-project.org/package=twitteR.
- 14.Abadi, M., Agarwal, A., Barham, P., Brevdo, E., Chen, Z., Citro, C., Corrado, G.S., Davis, A., Dean, J., Devin, M., Ghemawat, S.(2015). TensorFlow: Large-scale machine learning on heterogeneous systems. Software available from tensorflow.org.Google Scholar
- 15.Kingma, D. P., & Lei, J. Adam: A method for stochastic optimization. ICLR 2015.Google Scholar
- 17.Collobert, R., Weston, J., Bottou, L., Karlen, M., Kavukcuoglu, K., & Kuksa, P. (2011). Natural Language Processing (Almost) from Scratch. Journal of Machine Learning Research., 12, 2493–2537.Google Scholar
- 18.Mikolov, T., Chen, K., Corrado, G., & Dean, J. (2013). Efficient estimation of word representations in vector space, CoRR abs/1301.3781.Google Scholar
- 19.Cuzick, J. (2005). Rank Regression. In: Encyclopedia of Biostatistics (eds P. Armitage and T. Colton).Google Scholar
- 20.Hicham, Gueddah. “Introduction of the weight edition errors in the Levenshtein distance.” arXiv preprint arXiv:1208.4503(2012).