# Multiple Instance Learning Networks for Fine-Grained Sentiment Analysis

arXiv: https://arxiv.org/abs/1711.09645 · HTML: https://mlsanity.com/paper/1711.09645/

## Key facts

- Cited by 23 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 27 Nov 2017.

## Built on by (most cited first, 23 of 23)

- [Aspect-Controllable Opinion Summarization](https://mlsanity.com/paper/2109.03171/) (2021)
- [Multi-Vector Models with Textual Guidance for Fine-Grained Scientific Document Similarity](https://mlsanity.com/paper/2111.08366/) (2021)
- [Multi-Instance Causal Representation Learning for Instance Label Prediction and Out-of-Distribution Generalization](https://mlsanity.com/paper/2202.12570/) (2022)
- [SparseConvMIL: Sparse Convolutional Context-Aware Multiple Instance Learning for Whole Slide Image Classification](https://mlsanity.com/paper/2105.02726/) (2021)
- [WSDMS: Debunk Fake News via Weakly Supervised Detection of Misinforming Sentences with Contextualized Social Wisdom](https://mlsanity.com/paper/2310.16579/) (2023)
- [LLM-Enhanced Multiple Instance Learning for Joint Rumor and Stance Detection with Social Context Information](https://mlsanity.com/paper/2502.08888/) (2025)
- [A Role-Selected Sharing Network for Joint Machine-Human Chatting Handoff and Service Satisfaction Analysis](https://mlsanity.com/paper/2109.08412/) (2021)
- [A Multi-Grained Self-Interpretable Symbolic-Neural Model For Single/Multi-Labeled Text Classification](https://mlsanity.com/paper/2303.02860/) (2023)
- [Multiple Instance Neural Networks Based on Sparse Attention for Cancer Detection using T-cell Receptor Sequences](https://mlsanity.com/paper/2208.04524/) (2022)
- [A Speaker Turn-Aware Multi-Task Adversarial Network for Joint User Satisfaction Estimation and Sentiment Analysis](https://mlsanity.com/paper/2410.09556/) (2024)
- [W-RST: Towards a Weighted RST-style Discourse Framework](https://mlsanity.com/paper/2106.02658/) (2021)
- [Preliminary Steps Towards Federated Sentiment Classification](https://mlsanity.com/paper/2107.11956/) (2021)
- [A Discourse-level Multi-scale Prosodic Model for Fine-grained Emotion Analysis](https://mlsanity.com/paper/2309.11849/) (2023)
- [Text Classification using Graph Convolutional Networks: A Comprehensive Survey](https://mlsanity.com/paper/2410.09399/) (2024)
- [Comparative Snippet Generation](https://mlsanity.com/paper/2206.05473/) (2022)
- [Deep Multiple Instance Learning For Forecasting Stock Trends Using Financial News](https://mlsanity.com/paper/2206.14452/) (2022)
- [Improving Low-Resource Cross-lingual Parsing with Expected Statistic Regularization](https://mlsanity.com/paper/2210.09428/) (2022)
- [Large Discourse Treebanks from Scalable Distant Supervision](https://mlsanity.com/paper/2212.06038/) (2022)
- [Summative Student Course Review Tool Based on Machine Learning Sentiment Analysis to Enhance Life Science Feedback Efficacy](https://mlsanity.com/paper/2301.06173/) (2023)
- [FRACTAL: Fine-Grained Scoring from Aggregate Text Labels](https://mlsanity.com/paper/2404.04817/) (2024)
- [Detecting Scarce and Sparse Anomalous: Solving Dual Imbalance in Multi-Instance Learning](https://mlsanity.com/paper/2503.13562/) (2025)
- [milearn: A Python Package for Multi-Instance Machine Learning](https://mlsanity.com/paper/2512.01287/) (2025)
- [PREFER: Personalized Review Summarization with Online Preference Learning](https://mlsanity.com/paper/2605.05911/) (2026)

## Abstract (arXiv)

We consider the task of fine-grained sentiment analysis from the perspective of multiple instance learning (MIL). Our neural model is trained on document sentiment labels, and learns to predict the sentiment of text segments, i.e. sentences or elementary discourse units (EDUs), without segment-level supervision. We introduce an attention-based polarity scoring method for identifying positive and negative text snippets and a new dataset which we call SPOT (as shorthand for Segment-level POlariTy annotations) for evaluating MIL-style sentiment models like ours. Experimental results demonstrate superior performance against multiple baselines, whereas a judgement elicitation study shows that EDU-level opinion extraction produces more informative summaries than sentence-based alternatives.