Overview / Usage

This repository is about the sentiment analysis problem using SVM based ANN and feature extraction using Word2Vec model where Intel OneAPI toolkits is being integrated for efficient inference and performance

Methodology / Approach

Recent research suggests an emerging trend in the use of artificial intelligence to exceed human capabilities. One such application of AI is Sentiment Analysis, which shows potential in assessing the performance of products, marketing campaigns, and customer services. The analysis of consumer sentiment enables brands to comprehend their customers and improve their sales by developing strategies based on consumer reactions to various forms of content. Understanding consumer sentiment helps brands to tailor their messaging and advertising to appeal to their target audience, creating a personalized experience for each consumer and deepening their understanding of their customers. However, there is still ample room for improvement in Sentiment Analysis, especially in terms of enhancing its efficiency and effectiveness.

After experimenting with several optimizers for our model, we have determined that the SGD optimizer is the most suitable. This is due to its ability to converge to the minimum of the cost function, introduce random noise for regularization, and handle large datasets with efficiency. The results demonstrate that our model achieves the best performance when using the SGD optimizer, as illustrated in the figure.

Technologies Used

By utilizing Intel's OneDNN and OneMKL, we developed an application that utilizes the background operation optimization of OneDNN by fusing and reordering operations and reducing memory usage. We employed the Intel extension for PyTorch, which provided several benefits, like

1. Increased performance,
2. Improved efficiency,
3. Access to Intel-specific features,
4. Seamless integration, and
5. Flexibility. We observed a significant improvement in runtime and overall speed when using IPEX, as compared to the performance without using IPEX.

We also used Intel Extension for Scikit Learn and Modin: The drop-in replacement for Pandas which provided several benefits, like

1. Reduced Memory Usage
2. Improved Performance
3. Scalability
4. Access to Intel®-specific features like Math Kernel Library

Overall, the library optimizes data ingestion along with algorithmic computation to increase throughput and scalability

Repository

https://github.com/sammajum706/Intel_OneAPI_SentimentAnalysis

Other links

• Webapp (https://github.com/IamAnubhav-08/Intel-CodeMaven_OneAPI)