Multimodal Sentiment Analysis Project(Persian-3classes)

Overview

In this project, we explore multimodal sentiment analysis, which involves analyzing both text and image data together. Our goal is to predict sentiment labels (positive or negative) based on combined information from text and images.

Dataset Description

We obtained a dataset from Instagram that includes both text and image content.
The dataset consists of 1000 samples, each labeled as either positive or negative sentiment.
Text data is provided in an Excel file, while image data is stored in a RAR file.

Model Architecture

Our multimodal sentiment analysis model combines features from both modalities: Model

Image Modality:
- We use a pre-trained 16-layer VGG CNN to extract image features.
Text Modality:
- We employ a bidirectional LSTM (LSTM-BI) network to process text data.
The features extracted from both modalities are concatenated to form a joint representation.

Results

Our model achieved an accuracy of 69% on sentiment prediction.
Additional performance metrics and findings can be included here.

2024 4
2023 2
2021 2
2020 3

2024

Wav2Vec2 Sentiment Analysis Using Shemo Dataset

1 minute read

Overview In this project, we fine-tuned the Wav2Vec2 model to perform sentiment analysis based on both voice features and text transcripts from the Shemo da...

Psychological Health Chatbot:

1 minute read

Osmium Project Tasks

1 minute read

Overview In this project, we developed an Android application to estimate the location of cellular network cells using Received Signal Strength Indicator (RS...