Abstract This study investigates the comparative effectiveness of three modeling paradigms frequent pattern mining, Markov chains, and Long Short Term Memory (LSTM) networks in predicting user navigation behavior on the MSNBC Anonymous Web Navigation dataset. The dataset, comprising nearly one million anonymized browsing sessions, exhibits strong self transition dominance and low entropy, reflecting "sticky" user behavior where individuals tend to remain within a single content... This study investigates the comparative effectiveness of three modeling paradigms frequent pattern mining, Markov chains, and Long Short Term Memory (LSTM) networks in predicting user navigation behavior on the MSNBC Anonymous Web Navigation dataset. The dataset, comprising nearly one million anonymized browsing sessions, exhibits strong self transition dominance and low entropy, reflecting "sticky" user behavior where individuals tend to remain within a single content category rather than explore diverse sections of the site. Using next category prediction accuracy and perplexity as evaluation metrics, we find that while LSTM models achieve marginally higher performance, the improvement over first and higher order Markov models is not statistically significant (p>0.05). In contrast, frequent pattern mining despite its interpretability underperforms due to limited generalization beyond observed subsequences. Our results demonstrate that for structured, low entropy clickstream data, classical probabilistic models like Markov chains offer a compelling balance of predictive power, computational efficiency, robustness, and interpretability. These findings challenge the prevailing assumption that deep learning architectures inherently outperform simpler models in all sequential prediction tasks. Instead, they underscore the importance of aligning model complexity with the intrinsic characteristics of behavioral data a core tenet of domain driven actionable knowledge discovery. The study concludes that in many real world applications involving navigational or transactional logs, parsimonious models can deliver actionable insights without the overhead of deep neural networks, provided they respect the underlying behavioral dynamics. Future work may explore hybrid approaches or incorporate contextual signals to better capture rare but meaningful cross category transitions. Read More Read Less

Abstract This study introduces a dual dataset framework for analyzing the co-evolution of macro level digital infrastructure and micro level developer ecosystems. We integrate 45 years (1980-2020) of country level telecommunications indicators spanning mobile penetration, internet adoption, and broadband diffusion across 217 economies with a contemporary snapshot of 1,247 trending GitHub repositories, capturing realtime developer attention across 40+ programming languages. Through fork to star ratio... This study introduces a dual dataset framework for analyzing the co-evolution of macro level digital infrastructure and micro level developer ecosystems. We integrate 45 years (1980-2020) of country level telecommunications indicators spanning mobile penetration, internet adoption, and broadband diffusion across 217 economies with a contemporary snapshot of 1,247 trending GitHub repositories, capturing realtime developer attention across 40+ programming languages. Through fork to star ratio analysis, we identify distinct behavioral archetypes: research oriented languages (R, Jupyter Notebook) exhibit 4-13x higher forking rates (median ratios 0.29-0.38), reflecting academic reuse patterns for method adaptation and reproducibility, while infrastructure languages (Rust, TypeScript) demonstrate star-dominant engagement (ratios <0.04), indicating production tool consumption via stable APIs rather than source modification. Critically, we document the emergence of compositional language ecosystems where polyglot architectures deliberately stratify languages by computational layer Rust for systems safety, Python for AI orchestration, and TypeScript for interfaces moving beyond "language wars" toward purpose driven symbiosis. These patterns reveal programming languages function not merely as syntactic tools but as socio technical coordination mechanisms governing community behavior. The framework enables novel cross domain inquiries linking national infrastructure quality to global innovation patterns, with implications for digital policy design and open source ecosystem stewardship. Read More Read Less

Abstract This study investigates market anomalies in Amazon's stock price behavior through a 25-year analysis (2000-2025) of daily OHLCV data to identify high probability markers of fundamental revaluation episodes. Challenging the Efficient Market Hypothesis, the research examines two primary anomaly detection methodologies: identifying extreme volume spikes and quantifying overnight price gaps. Volume analysis revealed ten sessions with more than 100 million shares traded, with notable... This study investigates market anomalies in Amazon's stock price behavior through a 25-year analysis (2000-2025) of daily OHLCV data to identify high probability markers of fundamental revaluation episodes. Challenging the Efficient Market Hypothesis, the research examines two primary anomaly detection methodologies: identifying extreme volume spikes and quantifying overnight price gaps. Volume analysis revealed ten sessions with more than 100 million shares traded, with notable clustering in April 2007 and a concentration of events prior to 2010 reflecting Amazon's transition from speculative growth stock to established market leader. Price gap analysis identified ten sessions with absolute overnight gaps exceeding $9.90, predominantly occurring between 2021-2024 amid heightened macroeconomic volatility and sectorspecific pressures on technology valuations. The study integrates 20-day, 50-day, and 200-day simple moving averages to contextualize anomalies within hierarchical trend structures. Key findings demonstrate that extreme volume events serve as critical inflection markers where market consensus undergoes rapid recalibration, while overnight price gaps function as sentiment discontinuities frequently initiating multiday directional sequences. Confluence events where price crosses key moving averages on elevated volume- exhibited heightened predictive value for trend continuation or reversal. The research establishes that monitoring deviations from baseline volume and gap distributions provides early warning signals of potential trend transitions. Limitations include the need for catalyst attribution through correlation with news events and percentage based normalization to account for stock splits. This framework offers practitioners a robust methodology for distinguishing noise from signal in price behavior, with implications for algorithmic trading and risk management in volatile equity markets. Read More Read Less