India’s
conversational ai and chatbots industry,
in focus.

Grammar and Cognition Challenges

30.2% of this theme group

Researchers examine how tokenisation and language model design affect grammatical understanding and cognitive alignment.

Generative grammar (5) · Language model (5) · Cognition (4) · Grammar (3)

The Architecture of Linguistic Discretisation: A Comparative Analysis of Tokenisation Strategies in Large Language Models

Are you hallucinated? Insights into large language models

Low‑Resource Language Barriers

27.1% of this theme group

Studies highlight difficulties in translating and processing languages with limited data and tools.

Resource (3) · Natural language (5) · Translation (3) · Natural language processing (2)

Neural Machine Translation for Low-resource Languages: A Survey

A Systematic Review on Translation of Low Resource Languages Using Machine Translation System with Reference to Quality Estimation

Multilingual Chatbot Limitations

22.9% of this theme group

Works discuss obstacles in building chatbots that handle regional scripts and code‑switching.

Chatbot (4) · Machine translation (5) · Marathi (3) · Devanagari (3)

Bridging Languages in Healthcare: A Comprehensive Review of Multilingual and Code-Switched Chatbot Interactions

Sign Language Translation with Multilingual Support

Sentiment and Fake‑News Detection Issues

19.8% of this theme group

Analyses reveal challenges in extracting sentiment and identifying misinformation on social platforms.

Sentiment analysis (11) · Social media (7) · Fake news (3) · Systematic review (2)

Text Generation and Sentiment Analysis Using NLP

An Overview of Productive Sentiment Analysis with Various Feature Extraction Techniques Using Machine Learning Classifiers

Model Robustness and Interpretability

50.0% of this theme group

Authors caution about fragile fine‑tuning and the need for transparent language model assessments.

Language model (8) · Natural language (3) · Robustness (3) · Pattern recognition (2)

A Comparative Analysis of Learning-Rate Selection for Fine-Tuning Transformer Language Models

When measurement meets machine learning: interpretability and scalability in modelling item difficulty for language assessment

Bias in Ranking and Identification

27.5% of this theme group

Papers warn of systemic bias when ranking candidates or detecting fake content in regional languages.

Ranking (2) · Identifier (2) · Code (2) · Hindi (2)

Bias-Aware Resume Screening: A Survey and Analytical Framework Using NLP and Machine Learning

NLP-Based Investigation of Textual and Semantic Cues in Fake News Identification

Sentiment Cognition Boundaries

20.0% of this theme group

Research points to limits in sentiment categorization and the cognitive impact of AI‑generated text.

Sentiment analysis (3) · Cognition (2) · Categorization (2) · Social media (2)

Exploring the Boundaries of Reality: Investigating the Phenomenon of Artificial Intelligence Hallucination in Scientific Writing Through Cha

A Review of Aspect-Based Sentiment Analysis in Educational Contexts: From Transformers to Explainable AI

Vision‑Text Recognition Advances

22.3% of this theme group

Opportunities focus on combining pattern recognition with OCR and translation for multilingual document processing.

Pattern recognition (42) · Convolutional neural network (44) · Optical character recognition (18) · Machine translation (27)

A Hybrid Transformer–Classical Deep Learning OCR Framework for Multilingual Handwritten and Printed Document Recognition

Enhancing Speech Emotion Recognition with Deep Learning Techniques

Reliability through Knowledge Integration

22.1% of this theme group

Works explore enhancing language model consistency by integrating knowledge graphs and adaptive reasoning.

Similarity (21) · Natural language (35) · Reliability (13) · Consistency (15)

Hybrid Knowledge Graph Integration in Large Language Models for Explainable Contextual Understanding and Improved Accuracy

Controllable Epistemic Sensitivity in Large Language Models: Probing, Benchmarking, and Adaptive Reasoning

Scalable Generative Chatbot Frameworks

17.9% of this theme group

Studies propose multimodal architectures to improve chatbot coherence, pragmatics, and figurative language handling.

Chatbot (35) · Natural language (54) · Generative grammar (27) · Scalability (28)

Efficacy of Selected Generative AI Systems in Assessing Discourse Coherence, Pragmatic Competence, Collocational Competence, and Figurative

A Novel Multimodal Deep Learning Framework for Conversational <scp>AI</scp> : Integrating Vision, Text, and Speech With Knowledge‐Augmented

Social Media Sentiment for Markets

16.6% of this theme group

Research leverages sentiment analysis of online chatter to inform financial and market insights.

Sentiment analysis (170) · Social media (59) · Stock market (12) · Stock (10)

Sentiment analysis of twitter data using machine learning approaches and semantic analysis

Social Sentiment Insights Using Natural Language Processing

Misinformation Detection Techniques

10.8% of this theme group

Papers develop robust models for spotting fake news and harmful language in mixed‑language content.

Social media (61) · Fake news (14) · Misinformation (16) · Support vector machine (20)

Detecting Hate Speech in Code‐Mixed Languages: A Robust Study Leveraging Data Augmentation, Transfer Learning, and Model Explainability Tech

A Hybrid Deep Learning Framework for Detecting Harmful Language in Social Media Texts

Efficient Summarization Methods

5.8% of this theme group

Works aim to reduce redundancy and improve coherence in multi‑document and video summarization.

Automatic summarization (63) · Multi-document summarization (5) · Sentence (9) · Redundancy (4)

Graph-based Neural Multi-Document Summarization

AI-BASED EDUCATIONAL VIDEO SUMMARIZATION

Sign Language Recognition Progress

4.4% of this theme group

Research advances deep learning approaches for real‑time gesture and sign language interpretation.

Sign language (54) · Gesture (34) · Gesture recognition (23) · Sign (21)

Deepsign: Sign Language Detection and Recognition Using Deep Learning

Deep Learning-Based Real-Time Sign Language Recognition and Bi-Directional Communication Systems

Bengaluru

35.5% across the displayed hubs

19 companies hiring · $128.8M disclosed funding

Mumbai

13.9% across the displayed hubs

5 companies hiring · $23.8M disclosed funding

Delhi

13.9% across the displayed hubs

4 companies hiring · $1.4M disclosed funding

Hyderabad

8.9% across the displayed hubs

3 companies hiring · $6.6M disclosed funding

Chennai

8.2% across the displayed hubs

3 companies hiring · $765.0K disclosed funding

Pune

7.7% across the displayed hubs

2 companies hiring · $5.8M disclosed funding

Gurugram

6.4% across the displayed hubs

9 companies hiring · $26.7M disclosed funding

Noida

5.7% across the displayed hubs

4 companies hiring · $462.6K disclosed funding