Technical deep dives, research papers, and engineering logs from the Finsense team.
Is Dark Mode just for flex for developers or just an aesthetic for users? While the opinion may differ for various persons but for long term sessions like trading and code editor, dark mode is a necessity and not a fancy feature. In this blog, I have explained why dark mode is a necessity.
When I starts my studies for Deep Learning, all the shortforms I felt very puzzled. The two most that puzzled my studies is Convolutional Neural Network (CNN) and Recurrent Neural Network (RNN).
The whole internet is optimized for relevance and clicks but to get these relevance and clicks the search engines promote and prioritize positive storis and official press release and miss the negative stories that may have huge impact.
When the financial data is collected, it does not become useful instantly at that moment. The process of transforming raw market feeds into structured, reliable datasets using data pipelines is explained in this article.
Quantum computation's entry is considered to be the next breakthrough in the field of innovation. Scientists often believe that it is going be the next revolutionary leap in the information technology sector.
Overall accuracy is not enough to evaluate early predictive models in stock markets. The importance of recall and precision in the measurement of performance of the model is explored in this article along with their relevance on investment and trading decisions.
Now a days Deep Learning Models have become very crucial for high quality infrastructures from Face recognition to vehicle which are autonomous. The Robustness of the models against the malicious attack is very paramount. In this article We will investigate Adversarial Machine Learning (AML).
Artificial Intelligence is fundamentally reconstructing the financial landscape, driving changes from high-frequency trading algorithms to personalized banking experiences. This article delves into how AI technologies are enhancing efficiency, accuracy, and security across the financial sector.
A powerful tool in transforming unstructured market opinions and modern retail trading into actionable insights is sentiment analysis.
Large Language Models(LLMs) are very powerful models but they often gave a wrong result when they are tasked with something like Retrieval of Real Time Data or Queries which are Domain Specific.
While making the Finsense webapp we used a free zero api solution where we didn't use any costly api to fetch news rather we fetched news directly from the frontend but this had a major issue.
We all know that stock market daily operates on the concept of uncertainty. It utilizes on volatility and huge stream of data. To overcome the uncertainty, the concept of statistical data analysis has been introduced. Statistical analysis plays an important role in the modern stock market.
The whole financial world is based on one very important thing that is free and fair news. We aimed to provide our users with authentic financial news without the $24k Bloomberg API cost by using advanced scraping and Google News RSS.
Designing a website for a huge amount of public involves many challenges. This article focuses on technical and psychological foundations required to build a high traffic intensive interface.
In the past 60 years Multi-Layer Perceptron(MLP) has helped us as the fundamental block of deep learning based on the approximation theorem. Though MLP suffered from a critical lacking of interpretability and require massive parameter often. This article explores Kolmogorov-Arnold Network(KAN) model...
Due to scale differences and inconsistencies, analysis of financial data remains extremely difficult despite appearing in abundance. How raw financial data is transformed into reliable and inputs for meaningful and accurate analysis are explored in the given article.
The challenge of detecting tonal shifts in market news. Using transformer attention mechanisms to distinguish cynicism from optimism.
