[{"content":" Footnotes to a Mind # A growing archive of scientific and technical insights—curated from my weblog and beyond. This space is dedicated to ideas worth exploring, questions worth asking, and discoveries worth sharing. Whether you\u0026rsquo;re here to browse, learn, or dive deep, I hope these entries spark your curiosity and expand your perspective.\n","externalUrl":null,"permalink":"/resources/knowledge-base/","section":"Resources","summary":"A curated archive of scientific and technical insights, bringing together ideas worth exploring, questions worth asking, and discoveries worth sharing.","title":"Knowledge Base","type":"resources"},{"content":"Data science is an interdisciplinary field focused on extracting valuable insights from data. It merges methods from statistics, computer science, machine learning, and domain expertise to analyze and interpret complex datasets. The process begins with gathering and cleaning raw data, ensuring it\u0026rsquo;s suitable for analysis. Data scientists then use statistical models and algorithms to identify patterns and trends that can inform decision-making or predict future outcomes.\nApplications of data science span diverse industries, including healthcare, finance, marketing, and environmental science. For instance, it helps optimize business strategies, improve medical diagnoses, or predict consumer behavior. Advances in technology have made it possible to handle massive amounts of data, known as \u0026ldquo;big data,\u0026rdquo; further amplifying its potential.\nUltimately, data science is a powerful tool for solving real-world problems and driving innovation, playing a crucial role in today\u0026rsquo;s data-driven world.\n","externalUrl":null,"permalink":"/resources/knowledge-base/understanding-data-science/","section":"Resources","summary":"Data science is an interdisciplinary field focused on extracting valuable insights from data. It merges methods from statistics, computer science, machine learning, and domain expertise to analyze and interpret complex datasets …","title":"Understanding Data Science","type":"resources"},{"content":" Project: AI-Powered Price Optimization # The Challenge # A local retailer struggled with stagnant sales—their pricing strategy was stuck in the past!\nThe Solution # Using Python and price elasticity analysis, I built a dynamic pricing model that:\n📈 Increased total sales by $2,190 without changing inventory. 🎯 Personalized prices for 5 product segments (e.g., Beauty, Electronics). How It Works # Data Analysis: Uncovered trends in historical sales data. Elasticity Calculation: Identified which products could handle price changes. Dynamic Rules: Automated price adjustments (+5% for low-elasticity, -5% for high-elasticity). See the Magic\nExplore the code and datasets on GitHub\nThe Price Whisperer: How AI Boosted Sales by 2K in 30 Days! # Chapter 1: The Data Detective # It all began with a CSV file named retail_sales_dataset.csv—a treasure trove of 1,000 transactions. My mission? To crack the code of pricing chaos! Toolkit: Python, Pandas, Matplotlib. First Clue: Histograms revealed most products sold were in the $50-200 range. Plot Twist: Men bought expensive electronics; women preferred mid-range clothing!\nChapter 2: The Time Traveler # I transformed the Date column into a time machine! Weekly trends showed: Holiday Peaks: Sales spiked in November (Black Friday madness!). Spring Slump: Flat sales in March-April. Time to disrupt the status quo with dynamic pricing!\nChapter 3: The Elasticity Enigma # Price elasticity (elasticity = %ΔQty / %ΔPrice) was the key! Low-Price Products : Elasticity = -0.49 → \u0026ldquo;Raise prices, no one cares!\u0026rdquo; Luxury Products : Elasticity = +0.12 → \u0026ldquo;Cheaper = More prestige?!\nChapter 4: The Dynamic Pricing Revolution # I unleashed pricing rules like a wizard: Low-elasticity items: +5% price hike. High-elasticity luxuries: -5% discount. Result? Sales jumped from $373K to $375K—without losing customers!\n","externalUrl":null,"permalink":"/resources/projects/price-optimizing/","section":"Resources","summary":"A local retailer struggled with stagnant sales—their pricing strategy was stuck in the past!","title":"Price Optimizing","type":"resources"},{"content":" Project: AI-Driven Portfolio Optimization # The Challenge # An investment firm struggled with portfolio imbalance—their strategy couldn\u0026rsquo;t balance high returns with manageable risk!\nThe Solution # Using Python and Modern Portfolio Theory, I built an optimization engine that:\n🚀 Achieved 23% better Sharpe ratio than traditional portfolios\n📊 Simulated 10,000+ portfolio combinations to find the efficient frontier\nHow It Works # Market Alchemy: Analyzed 1 year of data for TSLA, GOOGL, AMZN, and BRK-B Risk-Reward Sorcery: Calculated volatility, correlations, and Sharpe ratios Optimal Blending: Automated weight allocation\nKey Discoveries # 🔥 GOOGL-AMZN show 0.52 correlation (higher risk/reward synergy)\n🛡️ BRK-B acts as portfolio stabilizer with low volatility\n📉 Golden Cross Detection: 50-day vs 200-day MA signals\nSee the Magic Explore the code and datasets on GitHub\nHow I Engineered a 23% Superior Portfolio Using Python \u0026amp; Modern Portfolio Theory # Chapter 1: Gathering the Ingredients # Every great potion begins with quality ingredients. I harvested:\n1 year of market data for TSLA, GOOGL, AMZN, and BRK-B\n50,000+ data points of OHLC prices and volumes\nPrecision tools: Pandas for data wrangling, yfinance for live feeds\nChapter 2: The Volatility Crucible # Raw data is wild magic - I tamed it with:\n50-day \u0026amp; 200-day Moving Averages revealing TSLA\u0026rsquo;s mood swings\nDaily return distributions showing AMZN\u0026rsquo;s kurtosis (fat tails)\nVolume spikes analysis predicting BRK-B\u0026rsquo;s quiet strength\nChapter 3: The Correlation Crystal Ball # I discovered hidden relationships through:\nHeatmap sorcery showing GOOGL-AMZN\u0026rsquo;s 0.52 correlation\nCovariance matrices proving BRK-B\u0026rsquo;s diversification power\nChapter 4: The Efficient Frontier Quest # 10,000 simulated portfolios revealed:\nMaximum Sharpe Ratio of 2.29 (23% better than S\u0026amp;P 500)\nOptimal weights: 17.8% AMZN, 5.4% TSLA, 15.3% GOOGL, 61.5% BRK-B\nMaximum Sharpe Return of 33%\nRisk-return sweet spot at 15% volatility\nChapter 5: The Golden Portfolios # Three elixirs of optimal performance emerged:\n18.5789 AMZN, 5.1455% TSLA, 15.8968% GOOGL, 60.3789% BRK-B\n16.6440%AMZN, 5.3499% TSLA, 14.6621% GOOGL, 63.3440% BRK-B\n17.8025% AMZN, 5.3034% TSLA, 15.3061% GOOGL, 61.5880% BRK-B\n","externalUrl":null,"permalink":"/resources/projects/stock-analysis-and-portfolio-optimization/","section":"Resources","summary":"AI-Driven Portfolio Optimization","title":"Stock Analysis and Portfolio Optimization","type":"resources"},{"content":" Project: AI-Driven Stock Market Forecasting # The Challenge # Predicting stock trends is like reading tea leaves—volatile and unpredictable!\nThe Solution # I built a machine learning-powered dashboard that:\n📊 Visualizes real-time data with interactive candlestick charts\n🔮 Predicts trends using Linear Regression\n📈 Identifies optimal buy/sell points with RSI \u0026amp; Bollinger Bands\nKey Insights # Strong Correlation: Google \u0026amp; Amazon stocks move together (0.95 correlation). Amazon\u0026rsquo;s Volatility: 12% daily swings require careful timing. ML Predictions: $1 Google ↑ → $0.95 Amazon ↑. Tech Stack # Python, yfinance, Plotly, Scikit-learn Live updates via Yahoo Finance API Explore the code and datasets on GitHub\nThe Data Alchemist: Transforming Raw Market Data into Actionable Insights # Act 1: Hunting Live Data # Live stock market data flows like a roaring river! Using the yfinance library:\nExtracted 3 years of historical data for Google (GOOG), Apple (AAPL), and Amazon (AMZN). First Insight: Amazon showed 12% higher volatility compared to others! Act 2: Technical Enchantments # I crafted two powerful spells to decode stock secrets:\nRelative Strength Index (RSI): # RSI \u0026gt; 70 → Overbought signal\nRSI \u0026lt; 30 → Oversold signal\nBollinger Bands: # Price touches upper band → Likely price drop\nPrice touches lower band → Likely price rally\nAct 3: The Interactive Crystal Ball # With Plotly, I built a magical dashboard:\nLive candlestick charts with zoom/pan functionality.\nBollinger Bands in bold colors.\nRSI displayed on a secondary axis.\nAct 4: Unlocking Hidden Relationships # The correlation matrix revealed:\nGoogle \u0026amp; Apple: 0.87 correlation\nGoogle \u0026amp; Amazon: 0.95 correlation → Twin trends!\nAmazon\u0026amp; Apple: 0.83 correlation\nAct 5: Machine Learning Prophecy # With linear regression, I predicted:\n$1 increase in Google → $0.95 rise in Amazon !\nMultivariate model achieved 82% accuracy in forecasting Apple’s trends.\n","externalUrl":null,"permalink":"/resources/projects/stock-market-analysis-using-python/","section":"Resources","summary":"AI-Driven Stock Market Forecasting","title":"Stock Market Analysis Using Python","type":"resources"},{"content":" Project: \u0026ldquo;The Viral Codebreaker: Decoding YouTube\u0026rsquo;s Trending Algorithm\u0026rdquo; # The Challenge # YouTube\u0026rsquo;s trending algorithm is a black box—creators waste millions trying to guess what works. How can we systematically identify the patterns behind viral videos?\nThe Solution # I built a YouTube analytics engine that:\n🚀 Analyzed 200+ trending videos across 15 categories\n🔍 Discovered hidden correlations between metadata and engagement\n📈 Boosted content strategy insights by 83% for creators\nHow It Works # API Wizardry: # Harvested trending videos using YouTube\u0026rsquo;s API\nData Alchemy: # Transformed raw JSON into actionable insights\nPattern Decryption: # Category dominance analysis\nOptimal video length detection\nEngagement correlation mapping\nKey Discoveries # ⏱️ Sweet Spot : 8-12 min videos get 2.3x more comments\n🎮 Gaming Dominance : 32% of trending content\n🏷️ Tag Magic : 7-11 tags optimize discoverability\n🕒 Prime Time : 2-5 PM PST uploads gain 40% more views\nMrBeast Case Study # \u0026ldquo;Applied these insights to YouTube\u0026rsquo;s king of viral content:\n📊 92% correlation between video length and view duration\n🎯 Strategic tagging accounted for 37% of recommendation traffic\n⏰ Prime-time uploads boosted first-hour views by 63%\nWhy This Matters? # Data-Driven Creativity: Replace guesswork with algorithm-backed strategies Platform Insights: Understand YouTube\u0026rsquo;s recommendation patterns Competitive Edge: Benchmark against category leaders\nExplore the Code Magic\nHow I Reverse-Engineered YouTube\u0026rsquo;s Secret Sauce Using Python \u0026amp; Data Science # Chapter 1: API Alchemy – Harvesting YouTube’s Secret Data # Like a digital Indiana Jones, I raided YouTube’s API temple:\nEngineered pagination logic to bypass 50-result limits, extracting 200+ trending videos\nCaptured 15+ metrics per video: from engagement stats to ISO 8601 duration codes\nBuilt fault-tolerant requests handling API quotas and sudden rate limits\nWhy it matters: Raw data is the ore – this chapter mined the motherlode.\nChapter 2: Data Forge – Transforming Chaos into Insights # Raw JSON is modern hieroglyphics – I became a data Rosetta Stone:\nISO 8601 Decoder: Transformed cryptic PT15M33S into plottable seconds\nTag Array Reshaper: Converted stringified \u0026ldquo;[list]\u0026rdquo; to actual Python lists\nTime Zone Wizardry: Localized Qatar time with pytz for accurate publish-hour analysis\nKey Insight: Cleaned data revealed 8-12 minute videos dominate trending pages.\nChapter 3: Visualization Cathedral – Painting Stories with Seaborn # Numbers are shy – I made them dance through 14+ visual types:\nHeatmap Revelation: Discovered 0.62 correlation between likes \u0026amp; comments\nDuration Binning: Exposed sweet spot\nPublish-Time Radar: Mapped Any country prime-time\nGolden Nugget: Word clouds exposed trending titles love “Video” and “Question”.\nChapter 4: The MrBeast Blueprint – Reverse-Engineering Virality # I cracked the code of YouTube’s Midas:\nDuration Pattern: 67.74% of his long video viral hits cluster at 10-20 min\nTag Strategy: 6.57 average tags/video (vs 14.27 for trending vidoes)\nUpload Chronurgy: 17 UTC= peak algorithm receptivity\n","externalUrl":null,"permalink":"/resources/projects/youtube-data-collection-and-analysis/","section":"Resources","summary":"The Viral Codebreaker: Decoding YouTube’s Trending Algorithm","title":"YouTube Data Collection and Analysis","type":"resources"},{"content":" Who am I? # The Journey So Far # Hello, and welcome! I’m SaMoA. With a master’s degree in quantum computation and a thesis on Randomness in Quantum Measurements, I began my journey exploring abstract mathematical concepts and computational systems. Today, I bring that same analytical mindset to the exciting field of data science. My transition into data science is rooted in a deep fascination with financial data and markets. I’m passionate about leveraging machine learning and statistical methods to uncover trends, predict outcomes, and solve real-world financial challenges. My unique combination of theoretical rigor and practical experience allows me to deliver impactful, data-driven solutions.\nAreas of Expertise # Data Science \u0026amp; Analysis: Proficient in data preprocessing, exploration, and visualization with Python. Financial Data Modeling: Skilled in analyzing market trends and creating predictive models for financial applications. Machine Learning: Developing algorithms and deploying models tailored to tasks like stock price forecasting and portfolio optimization. Mathematics \u0026amp; Computation: Leveraging a strong foundation in mathematics to address challenges involving uncertainty and complex datasets. Programming: Experienced in crafting efficient code and collaborating through platforms like GitHub and Google Colab. Get in touch at mailto:contact@samoaalamdari.com.\n","externalUrl":null,"permalink":"/about/","section":"Home Page","summary":"I’m interested in engineering, programming, research and science","title":"About","type":"page"},{"content":" To contact me email me at any time. Fell free to follow me on LinkedIn. Follow me on YouTube. Follow and star me on GitHub. ","externalUrl":null,"permalink":"/contact/","section":"Home Page","summary":"ّIn this page you’ll find out different ways to contact me.","title":"Contacts","type":"page"},{"content":" Scientific Background # Dive into my scientific journey. Got a favorite topic? I\u0026rsquo;d love to hear about it.\n","externalUrl":null,"permalink":"/resources/projects/","section":"Resources","summary":"","title":"Articles \u0026 Projects","type":"resources"},{"content":"","externalUrl":null,"permalink":"/authors/","section":"Authors","summary":"","title":"Authors","type":"authors"},{"content":"","externalUrl":null,"permalink":"/categories/","section":"Categories","summary":"","title":"Categories","type":"categories"},{"content":" Projects # Explore a curated selection of my work—from technical experiments to creative endeavors. Each project reflects my passion for problem-solving, innovation, and continuous learning.\nBackground # Get to know the journey behind the name. This page offers a glimpse into my academic path, professional experiences, and the values that shape my approach to work and life.\nWeblog # A space for deep dives into science and technology that spark my curiosity. From thought-provoking concepts to hands-on experiments, this is where I share insights, discoveries, and ideas that keep my mind buzzing.\n","externalUrl":null,"permalink":"/","section":"Home Page","summary":"","title":"Home Page","type":"page"},{"content":"","externalUrl":null,"permalink":"/resources/","section":"Resources","summary":"","title":"Resources","type":"resources"},{"content":"","externalUrl":null,"permalink":"/series/","section":"Series","summary":"","title":"Series","type":"series"},{"content":"","externalUrl":null,"permalink":"/tags/","section":"Tags","summary":"","title":"Tags","type":"tags"}]