Data Science

The first complete bibliometric evaluation of every Greek university department and its tenured faculty: 23 universities, 339 departments and 8,809 faculty members, ranked across three bibliometric sources.

Estimate honestly. Model fearlessly. This book turns T-shirt sizing into fast, team-friendly estimation and hands you a complete e-commerce warehouse build on BigQuery—star schemas, ETL pipelines, performance tuning and all. From decision frameworks to real-world trade-offs, you'll gain the confidence to plan realistically and architect data systems that scale.

From your first predictive model to the ensembles that win competitions — the machine-learning core of the Data Science, End to End series, built from first principles in clean, idiomatic Python.

Polars 2.0 changed defaults that don't raise errors: lazy queries now stream, and joins no longer keep row order. This short guide shows exactly what breaks when you upgrade from 1.x, with every example run under both 1.44.2 and 2.0.0 and the real output printed side by side.

Go from your first MATLAB script to professional engineering projects. Master programming, data analysis, visualization, numerical computing, and AI-assisted workflows through practical, project-based learning.

Build one real AI system end to end, from raw job postings to a deployed, tested API, and learn production AI engineering on the way. 24 chapters, runnable code, and numbers you can reproduce.

为什么参数不大的模型，把信息放对地方就能反超更大的模型？本书以"压缩"为主线，把大模型的表示、泛化、推理、RAG 与智能体统一成一道信息管理题：什么该压、什么不能丢、什么该外置、放在哪里、何时取回。18 章、190 张原创图解，从熵一路讲到世界模型。

Go beyond running MR code. Learn how to design, analyze, interpret, and report a Mendelian Randomization study—from GWAS data to a publication-ready manuscript. Built for graduate researchers in epidemiology, biostatistics, genetics, and public health.

The intelligence is in the data. Learn to build AI systems that answer with grounded confidence, not fluent invention — through vector databases, RAG, and semantic search, from first principles to production-ready blueprints.

Aprende a combinar SQL, Python, estadística y Machine Learning para priorizar posibles fraudes en retail. Un enfoque práctico, técnico y aplicado, con ejemplos, visualizaciones y códigos de apoyo.

Passage 1 — Chapter 9, "Data Security": a definition that sets the technical tone immediately A hacker is someone skilled at finding undocumented techniques and loopholes in the tangled architecture of complex information systems; by intent, a "white hat" looks for such loopholes in order to strengthen the system, while a "black hat" uses them to steal or to cause harm. Passage 2 — Chapter 14, "Metadata Management": an image that explains the whole topic in one paragraph A vivid illustration: an enormous document archive with no index at all — the shelves are full, but there is no way to find out what sits on them short of examining every single box by hand. That is exactly what an organization looks like when it has piled up mountains of data without also taking care of metadata — the information physically exists, but it cannot be used systematically, since the only way to find what is needed is to already know where it sits. Knowledge about data is always scattered: in a large company, one person carries the structure of a single database in their head, another the rules of a single integration, a third the history of a single metric, and nobody holds the complete picture. Passage 3 — Chapter 15, "Data Quality Management, Part 1": why "quality" is an empty word without a yardstick A postal address missing an apartment number works perfectly well for a mass catalog mailing and works terribly for a courier who has to knock on the right door — and in both cases it is the very same row in the very same database, only the yardstick applied to it differs. Passage 4 — Chapter 21, "Organizational Change Management": the book's closing summary Data governance, the coordinating hub of eleven knowledge areas this book opened with, stays an empty frame until specific people come to value the new way of working with data through their own experience — which is exactly why a book that began by mapping the circle of disciplines around that hub fittingly closes not with another technique or tool, but with a conversation about the person without whose deliberate participation no structure ever becomes a practice.

Passage 1 — Chapter 8, "Data Quality: Can the Board Trust Its Own Numbers": what bad data actually costs a leader Business folklore has long carried a caustic image: a machine that hands back garbage on the way out once it has been fed garbage on the way in — the old saying that poor input inevitably produces poor output, which this book already invoked earlier to explain why data quality went neglected for so long. It is worth taking that image seriously here and translating it into terms a leader can act on: exactly how much the input garbage costs a company, and by what test to recognize the moment a figure on a board slide becomes too risky to trust. Passage 2 — Chapter 9, "Analytics for Business: Data Warehousing and Metadata": a scene that stays with the reader Picture a board meeting: a single figure sits on the screen — quarterly revenue, say, or the share of customers who defected to a competitor — and one of the directors simply asks where that number came from. The pause that falls over the room while someone runs off to fetch an explanation from whoever put the report together says more about the state of trust in a company's analytics than any presentation ever could. Passage 3 — Chapter 11, "Data Management Maturity Assessment": maturity as something measured, not graded The practical conclusion for a leader is this: a maturity rung is not a verdict of "good" or "bad" — it measures how controlled and predictable a company's data work actually is. The higher the rung, the fewer surprises, and the more accurately a leader can forecast the consequences of decisions built on that data. Passage 4 — Chapter 13, "Managing Organizational Change in Data Management": the book's closing argument Taken together, the material across all thirteen chapters gives a leader a coherent view of data — from understanding it as a business asset to a concrete toolkit for leading organizational change. The architectural choices, technological infrastructure, and quality-control procedures examined earlier remain unrealized potential until someone at the most senior level of management personally takes on the work of carrying people through resistance and turning that potential into the company's everyday working habit.

This book presents a modern, unified, and language-agnostic workflow for creating analytical pipelines that are truly reproducible, from environment to execution.

Class imbalance isn’t a problem. Poor methodology is. This book challenges outdated practices and provides rigorous, data-driven alternatives. We focus on selecting the right tools, threshold tuning, real costs (not class frequencies), and strategic evaluation metrics, to build models that work.