Socio-Sentic framework for sustainable agricultural governance
Introduction
Social security is one of the imperative foundations for the overall growth and development of any nation. Several individual parameters and significant practices contribute in this direction of global progression. Amongst them, agriculture is one of the most potential channels for plummeting poverty and assuring livelihood security. It is the science of cultivating commodities necessary to nurture and prosper the human life. There has been a renaissance of interest in transforming agriculture into a long-term sustainable sector with the help of practical, plausible and innovative solutions. One of the significant aspects required for the sustainable development of agricultural sector is good governance and worthy allied policies. As governing bodies are getting more conscious while planning and implementing distinct agricultural schemes and policies, it becomes indispensable to endorse good governance. The United Nations Development Programme (UNDP) defined the characteristics of good governance as Participation; Rule of law; Transparency; Responsiveness; Consensus orientation; Equity; Effectiveness and efficiency; Accountability; and Strategic vision [1]. As a step to cultivate the good governance in agriculture with a long-term sustainability goal, this research proffers a technology-based solution to capture and analyse the public outreach and consensus on policies and procedures. The critical question here is that, how a particular agricultural policy will impact the life of a common man or society? Considering this, the cogitation of public opinion has become an essential step in the process of policy evaluation for its successful implementation & completion.
Social media has been embraced as an e-communication and e-participation real-time tool by government authorities. Its increased openness and transparency, global outreach and low-cost experience make it a powerful government barometer. Twitter is one of the most popular social networks worldwide and as per the statistics for the first quarter of 2018, this micro-blogging service averaged at 336 million monthly active users globally [2]. The platform is used as a communication channel by businesses, celebrities and governments. Gauging the public opinion for market and business intelligence is a well acknowledged domain of study with pertinent primary and secondary studies published in this direction of research [[3], [4], [5], [6]]. Mining public opinion from twitter datasets for government intelligence is a contemporary research practice. There is promising evidence on social media improving transparency of organisations and government ministries, but less evidence on whether this improves accountability. The accountability of government is a direct measure of its social responsibility and sustainability. Motivated by this, in this research, we propose a socio-sentic framework for sustainable agricultural governance using supervised learning based opinion mining on social web. The tweets associated with the Indian agricultural policy "Pradhan Mantri Fasal Bima Yojna" [7,8] has been evaluated to realize and validate the framework which extracts and analyzes public sentiments regarding the scheme. Pradhan Mantri Fasal Bima Yojna is one of the long awaited, path breaking scheme for Indian farmers has been constituted purposely with a view to provide financial aid to farmers. Due to diversifying and deviating climatic conditions of different regions in India, proper, protective and profitable crop production is a big challenge. Consequently, the scheme turned up with commendable features favouring the farmers in terms of insurance coverage, risk coverage (crop failure, post harvest losses), stabilisation of income and adopting latest agricultural technologies. Five supervised learning algorithms, namely, the Naive Bayes (NB), Support Vector Machine (SVM), Multilayer Perceptron (MLP), k-Nearest Neighbour (kNN) and Decision Trees (DT) have been empirically evaluated based on the classifier performance measures. The results specify the best sentiment classifier for predictive analysis of public opinion in tweets, on the government agricultural policies. As a step towards good governance for sustainable agriculture development, the proposed framework demonstrates the use of predictive learning model for government policy evaluation using opinion mining on social media.
The term ‘Socio-Sentic’ exhibits the two principal components of the framework. The word ‘Socio’ concerns to the ‘sociological’ or ‘society’ aspect of the framework and more specifically to the ‘social web’ which is used as a dataset for analyzing the publically available online content. The word ‘Sentic’ refers to implicit meaning/features associated with natural language exploited for tasks such as emotion recognition from text/speech or sentiment analysis [9]. Thus, the Socio-Sentic framework signifies the social web based framework for discovering patterns within the user-generated online content for typical text mining tasks such as opinion mining. Sentiment analysis or opinion mining is the computational study of people's opinions, appraisals, attitudes, and emotions toward entities, individuals, issues, events, topics and their attributes [3,10]. The sentiment polarity of a given text is classified into positive, negative or neutral categories. Twitter has been a widespread choice to perform sentiment analysis with its applications studied and reported across recent literature [6,10]. It serves as a goldmine for mining thoughts, opinions, expressions, emotions, reviews etc. which can be perpetually scrutinized for getting proper insights for market, business, and government intelligence. This research expounds the framework for government intelligence specifically for sustainable and accountable agricultural governance.
The rest of the paper is organized as follows: Section 2 introduces the proposed socio-sentic framework for sustainable agricultural governance to depict the relativity of good governance to sustainable agriculture. Section 3 provides an overview of ‘Pradhan Mantri Fasal Bima Yojana’ which is an agriculture-oriented service for societal profitability. Section 4 discusses the overall purpose of this work to validate the proposed framework by evaluating the chosen policy for sustainable agricultural governance using supervised learning approaches. It includes gathering and filtration of data, a brief summary about the famous supervised machine learning techniques used, statistical representation of data in a tabulated and graphical manner, and finally the sentiment classification of tweets for the sake of realizing public perception about the scheme. Section 5 summarizes the observations and calculations for comparison of the performance of different classifiers in terms of standard efficacy measures namely, precision, recall and accuracy. In the final section, section 6, the conclusions that have been drawn from the work undertaken along with the future work has been given.
Section snippets
Socio-Sentic framework for sustainable agricultural governance
Though much literary work has been done to establish the relation between good governance and agriculture for sustainability [11,12], no framework or technology-based solution has been provided so far. The government has, at the same time recognized the merits of social media and has begun to use them for informational, interactional, participatory and collaborative purposes. The official Twitter handle for Office of the Prime Minister of India, @PMOIndia with nearly 26.5 million followers is
Pradhan Mantri Fasal Bima Yojana – an agriculture based worldly welfare utility
The Agriculture sector has a pivotal role in Indian economy. It contributes to an approximate GDP of 15% and provides the principal means of livelihood for over 60 percent of Indian population. It plays an important role to fulfil the continuous food demand of growing population. Provision of raw material to industries support it to make a significant position in international trade. It is one of the largest employment providing sector in India which has a major contribution towards capital
Opinion mining for sustainable agricultural governance
The agriculture production and development dynamics have evolved over the past decades. The agricultural revolution has transformed from the basic crop farming towards a sustainable development avenue for smart agriculture. The advent of intelligent techniques has changed the landscape of conventional agriculture tactics. Different processing production and distribution strategies have now become an essential component of sustainable agriculture. Sustainable agriculture now acts as the lifeline
Results and analysis
This section highlights the results and observations related to performance of the proposed framework. Opinion mining has been performed as a text classification process by the applying supervised machine learning algorithms, namely, the Naive Bayes (NB), Support Vector Machine (SVM), Multilayer Perceptron (MLP), k Nearest Neighbour (kNN) and Decision Trees (DT), over a dataset consisting of 1008 tweets of PMFB yojana extracted from twitter. Precision, Recall and Accuracy have been used as
Conclusion
Sustainable agriculture encompasses security of food and fiber needs, natural resources, preserving environment, animal welfare, public health and much more to strengthen the socio-economic status, growth and development of a nation. It is vital to incorporate superior processes and practices in the field of agriculture to make it viable and valuable. Agricultural governance ensures apposite execution of this phenomenon by launching numerous schemes and policies. Although, the policies are
References (18)
- et al.
Sentiment analysis: a perspective on its past, present and future
Int. J. Intell. Syst. Appl.
(2012) - United Nations Development Programme (UNDP) policy document. Governance for sustainable human development, New York:...
- Social Media Statistics....
- et al.
Sentiment analysis on twitter
Int. J. Comput. Sci. Issues (IJCSI)
(2012) - et al.
Emotion Analysis Of Twitter Using Opinion Mining, Contemporary Computing (IC3)
(2015) - et al.
Opinion mining to assist user acceptance testing for open-beta versions
J. Inf. Assur. Secur.
(2017) - et al.
Proceedings of the World Congress on Engineering and Computer ScienceEmpirical Study of Twitter and Tumblr for Sentiment Analysis Using Soft Computing Techniquesvol. 12017
Empirical Study of Twitter and Tumblr for Sentiment Analysis Using Soft Computing Techniques
(2017) - Pradhan Mantri Fasal Bima Yojana, Ministry of Agriculture & Farmers Welfare....
- Crop Insurance....
Cited by (13)
A survey of the applications of text mining for agriculture
2019, Computers and Electronics in AgricultureCitation Excerpt :The approach proposed by Nimirthi et al. (2019) used a supervised approach that used CNNs to classify tweets. The aforementioned papers were seeking to identify opinion about agricultural in general, however, it is also possible to use the same techniques to estimate the popularity or effectiveness of government agricultural policy (Kumar and Sharma, 2018). They used a supervised machine learning strategy to classify tweets and infer public opinion about agricultural policy.
Sentiment Analysis for the Natural Environment: A Systematic Review
2023, ACM Computing SurveysWhat hurt the most? Assessing the influence of pollution and climate change on agricultural sustainability in Asia
2023, Sustainable DevelopmentTVET model for building safety and sustainable agriculture in Iran by 2030 and beyond
2023, Environmental Education ResearchArtificial intelligence research in agriculture: a review
2022, Online Information ReviewA novel game theory based reliable proof-of-stake consensus mechanism for blockchain
2022, Transactions on Emerging Telecommunications Technologies