Embedded Files
My Research Paper Summary
My Research Paper Summary
I have a diverse research background covering Machine Learning, Environmental Science, and Public Health. My completed works include a comparative study of data mining techniques for heart disease classification, an analysis of soil salinity in the Sundarbans to track climate change impacts, and a study on hospital admission trends at Kurmitola General Hospital to improve resource management. Building on this foundation, I am currently very active in my university research life. I am working on several new conference papers and a comprehensive journal publication, focusing on more advanced data science topics and larger datasets. My goal is to continue expanding my research portfolio through ongoing university projects and contribute meaningful findings to the global academic community.
I have a diverse research background covering Machine Learning, Environmental Science, and Public Health. My completed works include a comparative study of data mining techniques for heart disease classification, an analysis of soil salinity in the Sundarbans to track climate change impacts, and a study on hospital admission trends at Kurmitola General Hospital to improve resource management. Building on this foundation, I am currently very active in my university research life. I am working on several new conference papers and a comprehensive journal publication, focusing on more advanced data science topics and larger datasets. My goal is to continue expanding my research portfolio through ongoing university projects and contribute meaningful findings to the global academic community.
Conference papers
Conference papers
A Machine Learning Approach for Quantifying Sea-Level Rise: A Conservative Baseline Forecast through 2030 using Prophet Architecture.
A Machine Learning Approach for Quantifying Sea-Level Rise: A Conservative Baseline Forecast through 2030 using Prophet Architecture.
The Main Goal
The researchers wanted to create a "safe" and realistic prediction for sea-level rise. Instead of looking at the worst-case scenarios, they focused on a "Conservative Baseline" the minimum amount we should prepare for to keep coastal buildings and people safe.
How We Did It
Real Data: We used 32 years of high-quality satellite data from NASA (from 1993 to 2025).
Smart Tech: We used a machine learning tool from Facebook called Prophet. This tool is great because it can ignore "noise" (like small seasonal changes) and focus on the long-term trend.
Key Findings
2030 Prediction: The sea level is expected to rise by 9.59 mm by the year 2030.
Growth Rate: The sea is rising at a rate of about 1.9 mm per year.
Comparison: This prediction is lower than some other global reports (like the IPCC), which say it could rise by 3.5 to 4.5 mm per year. The researchers believe their number is a more practical "safe lower bound" for engineers to use when building.
Impact on Bangladesh
The study looked closely at how this rise affects the coast of Bangladesh:
Extreme Risk: Satkhira is at the highest risk because the land is very low, which will cause salt water to get into fresh water.
High Risk: Khulna and Bagerhat face high risks of flooding and damage to the Sundarbans mangrove forest.
The Bottom Line
The researchers suggest that engineers in Bangladesh should use this 1.9 mm/year rise as the absolute minimum standard when building or strengthening coastal dams (polders) to protect the millions of people living in these areas.
ANALYSIS OF SOIL SALINITY AND EARTH METAL CONCENTRATIONS ALONG RIVER DISTANCE GRADIENTS IN THE SUNDARBANS.
ANALYSIS OF SOIL SALINITY AND EARTH METAL CONCENTRATIONS ALONG RIVER DISTANCE GRADIENTS IN THE SUNDARBANS.
The Secret Chemistry of the Sundarbans: How Distance from the River Changes Everything
The Secret Chemistry of the Sundarbans: How Distance from the River Changes Everything
The Sundarbans in Bangladesh is the world’s largest continuous mangrove forest, a delicate place where fresh river water meets salty sea tides. But did you know that how far a patch of land is from the riverbank completely changes its "personality"? A recent study analyzed 110 sample plots to understand this "river distance gradient."
Key Findings from the Forest Floor
Key Findings from the Forest Floor
The research discovered that the river acts like a dual delivery system it brings both stress and life to the soil.
Salinity (The Salt Stress): Soil is saltiest right next to the river because of frequent tidal flooding. As you move deeper into the forest (further from the river), the salt levels drop significantly.
Nutrients (The Plant Food): Interestingly, the river is also a source of fertility. Key nutrients like Ammonium, Phosphorus, and Potassium are found in much higher amounts near the riverbanks. The forest interior, while less salty, actually has fewer nutrients.
pH Levels: The soil near the river is slightly more alkaline (higher pH), while the soil further inland tends to become more acidic.
Metal Mystery (Zinc vs. Iron): The study found that Zinc is higher near the river, suggesting it is carried in by the water. However, Iron levels actually increase further inland because Iron comes from the soil's natural parent rock rather than the tides.
Why This Matters for Conservation
Why This Matters for Conservation
This "trade-off"—having plenty of food but high salt near the river versus low salt but less food inland—is a major challenge for mangrove trees.
With climate change and rising sea levels, there is a risk that the fresher "refuge" zones deep in the forest will become too salty. This data helps conservationists choose the right tree species for reforestation based on exactly how far they are being planted from the water.
Comparative Analysis of Classification Techniques in Data Mining: K-Nearest Neighbors, Naive Bayes, and Decision Trees based Approach.
Comparative Analysis of Classification Techniques in Data Mining: K-Nearest Neighbors, Naive Bayes, and Decision Trees based Approach.
Which Computer Method is Best for Predicting Heart Disease?
Which Computer Method is Best for Predicting Heart Disease?
In our digital world, we have a lot of health data. This research looked at three famous ways to use that data to find out if a person has heart disease. The study used records from 303 patients and looked at 14 details, such as their age, blood pressure, and cholesterol levels.
The Three Methods We Tested
The Three Methods We Tested
K-Nearest Neighbors (KNN): This method is simple and works by looking at patients with similar health results.
Naive Bayes (NB): This is a very fast method that uses math to guess the chance of a disease.
Decision Trees (DT): This method creates a list of "if-then" rules, like a flowchart, which makes it very easy for doctors to understand why the computer made a choice.
What We Learned
What We Learned
After testing all three, the KNN method was the overall winner for this heart data. It was the most accurate and, most importantly, it caught every single case of heart disease in the test (this is called 100% Recall).
However, Naive Bayes was the best at making sure it didn't wrongly label a healthy person as sick. Decision Trees were the least accurate in this test, but they are still very useful because they are the easiest for humans to read and trust.
The Bottom Line
The Bottom Line
There is no "perfect" method for every job. If you want to make sure you never miss a sick patient, KNN is the best choice. But if a doctor needs to see exactly how the computer decided, a Decision Tree is better. In the future, the researchers want to test even more advanced methods to help doctors save lives.
Monthly Hospital Admissions and Mortality Trends at Kurmitola General Hospital
Monthly Hospital Admissions and Mortality Trends at Kurmitola General Hospital
Hospital Trends at Kurmitola General Hospital: What the Data Shows
Hospital Trends at Kurmitola General Hospital: What the Data Shows
A recent study at Kurmitola General Hospital in Dhaka has revealed some very important patterns regarding when people get sick and visit the hospital. By looking at the data from an entire year, researchers found that hospital life changes significantly with the seasons.
Key Findings from the Study
Key Findings from the Study
Quiet Beginnings: The early months of the year (January to April) are usually the quietest, with only 8 to 58 admissions per month and zero reported deaths.
The Big Spike: Starting in May, the number of patients begins to climb. This leads to a massive peak in October, where admissions soared to over 470 cases.
Mortality Trends: While the overall death rate remained low throughout the year, it did increase slightly during the busiest months. The highest number of deaths occurred in November, with a total of 7.
Why This Matters
Why This Matters
The data suggests that there is a clear seasonal pattern—possibly linked to the winter season—that drives more people to the hospital late in the year. When admissions suddenly jump, it puts a lot of pressure on the doctors, nurses, and hospital resources.
Moving Forward
Moving Forward
To save more lives and provide better care, the study concludes that hospital administrators need to plan ahead. By preparing more beds, staff, and medicine specifically for the peak months of October and November, the hospital can better handle these surges.
Ultimately, this data reminds us that every "number" in a report represents a real person and a family in need of help.
Journal paper
Journal paper
Social Media
Social Media
Page updated
Google Sites
Report abuse