Identification and analysis of Indian farmers’ behavior toward adoption of new farming technologies and e-agriculture schemes through Twitter

Data label category	Count	Interpretation
Opinion	461	Positive, negative, or neutral
Praising	293	Positive sentiments
Queries	246	Questions and doubts
Undefined	205	Unable to determine or irrelevant
Hybrid	185	Multiple categories
Total	1390

Table 2:

Polarity score for some tweets

Clean text	Polarity_score			Replaced value
Clean text	Negative	Neutral	Positive	Replaced value
Educate heavy machinery operator farm constrc…	0.044	0.393	0.962	1
Dream best initiative innovative	0.000	0.099	0.901	1
Want house build house make happen	0.000	0.794	0.206	0
Research publish recently also show cool effect…	0.000	0.840	0.160	0
Say nutrition comparison ground grow food conc…	0.000	0.805	0.195	0

Table 3:

Model results using TF-IDF

Model	Accuracy	Precision	Recall	F1-score
LR	0.91	0.91	0.91	0.90
RFC	0.86	0.86	0.86	0.85
SVC	0.83	0.81	0.83	0.78
Multinomial NB	0.81	0.78	0.81	0.76
Extra Trees Classifier	0.86	0.86	0.86	0.86
Decision Tree Classifier	0.84	0.79	0.82	0.84

[i] LR, logistic regression; NB, Naïve Bayes; RFC, random forest classifier; SVC, support vector classifier; TF-IDF, term frequency-inverse document frequency.

[ii] These results reflect that the model is effective for understanding sentiments in tweets, with TF-IDF + LR showing the best performance.

[iii] The model’s performance is measured using the following metrics:

[iv] Accuracy (91%) → 91% of all predictions (both positive and negative sentiments) were correct.

[v] Precision (91%) → When the model predicted a sentiment (e.g., positive), it was correct 91% of the time.

[vi] Recall (91%) → The model correctly identified 91% of actual positive sentiment cases.

[vii] F1-score (90%) → A balance between precision and recall, indicating strong overall performance.

[viii] The TF-IDF + LR combination is highly effective in analysing user sentiment (e.g., Farmers feedbacks on adopting technologies and schemes, customer reviews, or discussions on agricultural topics).

[ix] This approach can help in understanding user opinions, improving services, and addressing concerns in the e-agriculture domain.

Table 4:

Model results using count vectorizer

Model	Accuracy	Precision	Recall	F1-score
LR	0.87	0.88	0.87	0.85
RFC	0.84	0.84	0.84	0.81
SVC	0.82	0.85	0.82	0.77
Multinomial NB	0.79	0.78	0.79	0.70
Extra Trees Classifier	0.87	0.88	0.87	0.86
Decision Tree Classifier	0.82	0.81	0.82	0.82

[i] LR, logistic regression; NB, Naïve Bayes; RFC, random forest classifier; SVC, support vector classifier.

Table 5:

Model results using FastText

Model	Accuracy	Precision	Recall	F1-score
LR	0.85	0.85	0.85	0.83
RFC	0.81	0.82	0.81	0.77
SVC	0.82	0.82	0.82	0.79
Multinomial NB	0.81	0.80	0.81	0.78
Extra Trees Classifier	0.82	0.83	0.82	0.77
Decision Tree Classifier	0.74	0.75	0.74	0.75

[i] LR, logistic regression; NB, Naïve Bayes; RFC, random forest classifier; SVC, support vector classifier.

Table 6:

Deep learning model results using FastText

Model	Accuracy	Precision	Recall	F1-score
LSTM FastText	0.76	0.74	0.76	0.74
GRU FastText	0.77	0.73	0.77	0.74

[i] GRU, gated recurrent unit; LSTM, long short-term memory.

Primary challenges in the adoption of new farming technologies.

Primary challenges in the adoption of farming schemes.

Farmers’ rating in the adoption of new farming technology.

Farmers’ rating in the adoption of farming schemes.

Table 7:

Label of tweets and their polarity

S. no.	Tweets	Label	Polarity
1.	PMKISAN scheme aimed at providing direct income support to farmers has ensured support for their sustenance.	Opinion	Positive
2.	Farmers highlighting the importance of organic farming, soil health, and reducing chemical inputs.	Opinion	Positive
3.	Irrigation & insurance have both seen huge improvement through agriculture schemes.	Praising	Positive
4.	The PMFBY has provided farmers with financial security, mitigating risks associated with crop loss.	Praising	Positive
5.	Farmers are committing suicide due to water scarcity, and individuals like you are directly responsible for the deaths of farmers in Maharashtra.	Opinion	Negative
6.	Not all farmers in India know about the agricultural schemes launched by the GoI. As a result, most farmers do not avail themselves of the benefits of these schemes.	Opinion	Negative
7.	What is crop insurance?	Query	Neutral
8.	What is Pradhan Mantri Kisan Samman Nidhi?	Query	Neutral
9.	Organic farming combines with precision agriculture, using drone monitoring to enhance yields while maintaining eco-friendly practices.	Hybrid	Positive
10.	Sustainable agriculture is not only about safeguarding the planet—it’s also about increasing farmers’ profitability by using fewer resources.	Hybrid	Positive
11.	Agriculture is changing, and we must prepare for it.	Undefine	Neutral
12.	Greater innovation in farming is essential to address major challenges.	Undefine	Neutral

[i] GoI, Government of India; PMFBY, Pradhan Mantri Fasal Bima Yojana.

Category	Subcategory	Topics		No. of reviews
General information	Farm size	Small (<5 acres)		286
	Farm size	Medium (5–15 acres)		193
	Large (>15 acres)		12
	Type of farming	Crop farming (e.g., grains, vegetables, fruits)		372
		Livestock farming (e.g., cattle, poultry, sheep)		73
		Mixed farming (both crops and livestock)		445
Horticulture (e.g., fruits, vegetables, flowers)		23
Aquaculture (fish or shellfish farming)		16
Dairy farming		03
Organic farming		04
Technology awareness and usage		Which of the following technologies are you familiar with?	Precision agriculture tools (GPS-guided equipment)		0
	Drones for monitoring crops		0
	IoT sensors (for soil, water, or weather monitoring)		3
	Automated machinery		403
	AI or data analytics		0
	Farm management software		0
	Which of the following technologies do you currently use on your farm?		Precision agriculture tools (GPS-guided equipment)		0
Drones for monitoring crops			0
IoT sensors (for soil, water, or weather monitoring)		0
Automated machinery		403
AI or data analytics		0
Farm management software		0
Challenges in adopting new technology		What are the main reasons you haven’t adopted new farming technology?	High initial costs		348
			Lack of technical knowledge		386
	Limited access to training		323
	Poor Internet or GPS access		303
	Concerns about data privacy		286
	Cultural resistance or tradition		243
	Uncertain return on investment		403
	Need for skilled labor		456
	Rate the following challenges in adopting new technology on a scale of 1–5 (1 = not a challenge, 5 = major challenge):		High initial costs	1	39
				2	49
		3		69
		4	88
		5	103
		Lack of technical knowledge	1	119
			2	85
			3	78
		4	101
		5	203
		Poor Internet or GPS access	1	176
2			63
3	82
4	74
5	94
Need for skilled labor	1	78
	2	73
	3	28
4	46
5	231
How important is it to have access to training and support when adopting new technologies?	Very important		437
	Important		26
	Neutral		12
Not very important		05
Not important at all		11
Financial and economic considerations	Do you believe the costs associated with new technology are a barrier to adoption?	Yes		443
Financial and economic considerations		No		29
Maybe		19

[i] AI, artificial intelligence.

Category	Sub-category	Topics		No. of reviews
General information	Farm size	Small (<50 acres)		286
	Farm size	Medium (50–500 acres)		193
	Large (>500 acres)		12
	Type of farming	Crop farming (e.g., grains, vegetables, fruits)		372
		Livestock farming (e.g., cattle, poultry, sheep)		73
		Mixed farming (both crops and livestock)		445
Horticulture (e.g., fruits, vegetables, flowers)		23
Aquaculture (fish or shellfish farming)		16
Dairy farming		03
Organic farming		04
Awareness and access to new farming schemes		Are you aware of government farming schemes available in your region?	Yes		323
	No		168
	Where do you typically get information on new farming schemes?	Local Government Offices		29
		KVKs		58
		Agricultural shows or expos		36
	Radio/TV programs		43
	Internet/social media		41
	Other farmers		116
	Which types of schemes are you most interested in?	Subsidies for seeds and fertilizers		231
Irrigation and water management schemes		48
Organic farming programs		23
Crop insurance schemes		261
Credit and loan programs		36
Technology adoption schemes (e.g., farm machinery, digital tools)		238
Soil health and conservation programs		196
Challenges in adopting new farming schemes		What challenges do you face in adopting new farming schemes?	Lack of information about schemes		143
			Complex application processes		186
			High costs involved		36
	Delayed payments or subsidy disbursements		217
	Limited access to financial institutions		183
	Difficulty in meeting eligibility criteria		53
	Language barriers in documentation		235
	Lack of trust in government programs		56
	Inadequate support or training		248
	Poor infrastructure (e.g., transportation, power supply)		212
	Rate the following challenges in adopting the new scheme on a scale of 1–5 (1 = not a challenge, 5 = major challenge):		High costs involved	1	196
				2	33
		3		28
		4	39
		5	27
		Lack of information about schemes	1	224
			2	28
			3	23
		4	36
		5	12
		Delayed payments or subsidy disbursements	1	14
			2	18
			3	16
4		63
5		212
Language barriers		1	36
		2	16
		3	17
4	19
5	235
Lack of trust in government programs	1	201
	2	40
	3	12
4	14
5	56
Poor infrastructure	1	23
	2	21
	3	28
4	39
5	212
Financial and economic challenges	Do you face financial challenges in accessing new farming schemes?	Yes		249
	No		74
	If yes, what are the main financial challenges?	Lack of access to credit or loans		203
	If yes, what are the main financial challenges?	High interest rates on loans		226
Delayed loan approvals		198
High initial costs of adoption		215
Support and training needs	How important is training in adopting new farming schemes?	Very important		296
Support and training needs	How important is training in adopting new farming schemes?	Important neutral		14
Not very important		08
Not important at all		05

[i] KVKs, Krishi Vigyan Kendra’s.

References

Authors

Metrics

Articles in this issue

DOI: https://doi.org/10.2478/ijssis-2025-0019 | Journal eISSN: 1178-5608

Journal RSS Feed

Language: English

Submitted on: Nov 9, 2024

Published on: May 16, 2025

Published by: International Journal on Smart Sensing and Intelligent Systems

In partnership with: Paradigm Publishing Services

Publication frequency: Volume open

Keywords:

scheme,

TF-IDF,

Related subjects:

Introductions and overviews,

Engineering, other

© 2025 Santosh Kumar Rai, Jyoti Prakash Singh, published by International Journal on Smart Sensing and Intelligent Systems
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License.

Volume 18 (2025): Issue 1 (January 2025)

Identification and analysis of Indian farmers’ behavior toward adoption of new farming technologies and e-agriculture schemes through Twitter

Figures & Tables

Table 1:

Table 2:

Figure 1:

Table 3:

Table 4:

Table 5:

Table 6:

Figure 2:

Figure 3:

Figure 4:

Figure 5:

Figure 6:

Figure 7:

Figure 8:

Table 7:

Figure 9:

Paradigm

My account