Papers by erica jayasundera
IEEE Software , 2025
Accurate prediction of RNA tertiary structure from sequence remains a critical challenge in compu... more Accurate prediction of RNA tertiary structure from sequence remains a critical challenge in computational biology. This research explores a hierarchical approach, emphasizing the crucial role of secondary structure prediction as an intermediary step. By accurately identifying secondary structure elements like stem-loops and hairpins, we aim to improve the precision and efficiency of subsequent tertiary structure modeling. We evaluate various machine learning models for secondary structure prediction and investigate their integration with tertiary structure prediction algorithms, demonstrating the potential for significant improvements in overall structural accuracy.
The digital landscape is rapidly evolving, with Artificial Intelligence (AI) playing an increasin... more The digital landscape is rapidly evolving, with Artificial Intelligence (AI) playing an increasingly prominent role. However, this growing reliance on AI introduces new vulnerabilities, as highlighted by the recent discovery of the "ConfusedPilot" attack. Researchers at the University of Texas at Austin's Spark Lab, led by Professor Mohit Tiwari, identified this novel cyberattack method targeting Retrieval-Augmented Generation (RAG) based AI systems.
The world of software engineering is a vast landscape, teeming with talented individuals. But wit... more The world of software engineering is a vast landscape, teeming with talented individuals. But within this realm exists a coveted tier, the top 1%the architects, the visionaries, the code wizards whose skills elevate them to an almost mythical status. What separates these elite engineers from the rest? This article delves into that question, offering a roadmap for aspiring programmers who yearn to reach that pinnacle. Beyond the Buzzwords: Defining Programmer Levels Dissecting the concept of the "top 1%" requires a more objective measure. Here, we introduce a programmer competency scale (0.0 to 3.0) to navigate this complex terrain.
Developing ML based predictive systems with worked example for Credit Customer Churn Prediction, 2023
Customer churn prediction is a critical task for businesses that rely on customer retention. By a... more Customer churn prediction is a critical task for businesses that rely on customer retention. By accurately predicting which customers are likely to churn, businesses can take proactive measures to retain them. This can lead to significant cost savings and increased revenue. Machine learning (ML) is a powerful tool for customer churn prediction. ML algorithms can learn from historical data to identify patterns and relationships that can be used to predict future churn.
1. Data collection
2. Data pre-processing
3. Feature engineering
4. Model training
5. Model evaluation
6. Model deployment
Thesis Chapters by erica jayasundera
An AI Framework for Improved Learning of Dynamic Models from Time Series Data (with Application t... more An AI Framework for Improved Learning of Dynamic Models from Time Series Data (with Application to Stock Price Prediction)
Talk on Challenges in the Mobile Application security and
vulnerabilities of mobile apps. There a... more Talk on Challenges in the Mobile Application security and
vulnerabilities of mobile apps. There are many possible
weaknesses within the Android Based Devices. In this case study we look at 25 vulnerabilities and how they occur. We also look at the remedial action we could take to overcome the vulnerabilities so that application can be secure for use.
case study outline :
V1: Architecture, Design and Threat Modelling
V2: Data Storage and Privacy
V3: Cryptography Verification
V4: Authentication and Session Management
V5: Network Communication
V6: Environmental Interaction
V7: Code Quality and Build Settings
V8: Resiliency Against Reverse Engineering
In this case study we would use two datasets and identify the
common subset of information so tha... more In this case study we would use two datasets and identify the
common subset of information so that we could present as an evidences of crime related
incidents overlapping the taxi traffic in particular area.
using Weka, Python folio map visualization , Juypter for data analysis
Drafts by erica jayasundera
Both Gemini and ChatGPT 4 are behemoths in the realm of generative AI, capable of feats previousl... more Both Gemini and ChatGPT 4 are behemoths in the realm of generative AI, capable of feats previously unimaginable. While they share the same core functionality of text generation, translation, and creative content creation, their strengths and weaknesses differ, making them suitable for varying purposes.
Books by erica jayasundera
Talks by erica jayasundera
Uploads
Papers by erica jayasundera
1. Data collection
2. Data pre-processing
3. Feature engineering
4. Model training
5. Model evaluation
6. Model deployment
Thesis Chapters by erica jayasundera
vulnerabilities of mobile apps. There are many possible
weaknesses within the Android Based Devices. In this case study we look at 25 vulnerabilities and how they occur. We also look at the remedial action we could take to overcome the vulnerabilities so that application can be secure for use.
case study outline :
V1: Architecture, Design and Threat Modelling
V2: Data Storage and Privacy
V3: Cryptography Verification
V4: Authentication and Session Management
V5: Network Communication
V6: Environmental Interaction
V7: Code Quality and Build Settings
V8: Resiliency Against Reverse Engineering
common subset of information so that we could present as an evidences of crime related
incidents overlapping the taxi traffic in particular area.
using Weka, Python folio map visualization , Juypter for data analysis
Drafts by erica jayasundera
Books by erica jayasundera
Talks by erica jayasundera