Capitalizing Data Science: A Guide to Unlocking the Power of Data for Your Business and Products
- Length: 254 pages
- Edition: 1
- Language: English
- Publisher: BPB Publications
- Publication Date: 2022-12-03
- ISBN-10: 9355511582
- ISBN-13: 9789355511584
- Sales Rank: #0 (See Top 100 Books)
Unlock the Potential of Data Science and Machine Learning to Your Business and Organization
Key Features
- Includes today’s most popular applications powered by data science and machine learning technology.
- A solid primer on the entire data science lifecycle, detailed with examples.
- An integrated approach to demonstrating the use of Image Processing, Natural Language Processing, and Neural Networks in business.
Description
Can you foresee how your company and its products will benefit from data science? How can the results of using AI and ML in business be tracked and questioned? Do questions like ‘how do you build a data science team?’ keep popping into your head?
All these strategic concerns and challenges are addressed in this book.
Firstly, the book explores the evolution of decision-making based on empirical evidence. The book then helps compare the data-supported era with the current data-led era. It also discusses how to successfully run a data science project, the lifecycle of a data science project, and what it looks like. The book dives fairly in-depth into various today’s data-led applications, highlights example datasets, discusses obstacles, and explains machine learning models and algorithms intuitively.
This book covers structural and organizational considerations for making a data science team. The book helps recommend the use of optimal data science organization structure based on the company’s level of development. Finally, the book explains data science’s effects on businesses by assisting technological leaders.
What you will learn
- Learn the entire data science lifecycle and become fluent in each phase.
- Discover the world of supervised and unsupervised learning applications and structured and unstructured datasets.
- Discuss NLP’s function, its potential, and the application of well-known methods like BERT and GPT3.
- Explain practical applications like automatic captioning, machine translation, and emotion recognition.
- Provide a framework for evaluating your team’s data science skills and resources.
Who this book is for
Startups, investors, small businesses, product management teams, CxO and all developing businesses desiring to leverage a data science team to gain the most from this book. The book also discusses the potential of practical applications of machine learning and AI for the future of businesses in banking and e-commerce.
Cover Page Title Page Copyright Page Dedication Page About the Author About the Reviewer Acknowledgement Preface Errata Table of Contents 1. Data-Driven Decisions from Beginning to Now Introduction Data-driven decisions and their phases Human-led and data-supported decisions Data-led and human-guided Applications of data science Initiation Acquisition Maintenance and expansion Retention Exit Regain Challenges that need to be solved Conclusion 2. Data Science Life Cycle —Part 1 Introduction What is considered a data science project? Explanatory Predictive The key players The stages of a data science project Phase 1—The business problem phase Phase 2—The math problem phase Ownership of the overall solution Problem understanding stage Key stakeholders and accountability Consulting stage Typical outputs of this stage Key stakeholders and accountability Solution blueprint Power ABC Key stakeholders and accountability Typical outputs of this stage Finalizing success criteria Typical outputs of this stage Key stakeholders and accountability Finalizing reporting requirements Key stakeholders and accountability Implementation decisions Typical outputs of this stage Key stakeholders and accountability Roll-out roadmap Typical outputs of this stage Key stakeholders and accountability Conclusion 3. Data Science Life Cycle —Part 2 Introduction Data understanding Key stakeholders and accountability Typical outputs of this stage Data validation Accuracy Availability Completeness Reliability Key stakeholders and accountability Typical outputs of this stage Algorithmic solution Training and validation Data processing Missing value treatment Outlier treatment Variable or feature transformations Feature selection Algorithm development Validation Estimating business impact Key stakeholders and accountability Typical Outputs of this Stage Data and model governance Technical robustness Data robustness Compliance criteria Experimentation methodology Model governance teams Key stakeholders and accountability Typical outputs of this stage Model deployment and go-live Role of ML engineers Latency Throughput Memory CPU Network latencies Latencies across different steps Alerting and monitoring mechanisms Key stakeholders and accountability Typical outputs of this stage Measurement of production results Key stakeholders and accountability Typical outputs of this stage Optimization Optimizing algorithms Key stakeholders and accountability Typical outputs of this stage Roll-out Key stakeholders and accountability Typical outputs of this stage Conclusion 4. Deep Dive into AI Introduction Difference between AI, data science, and machine learning Ux& Interfaces Experimenter Data collectors/storage Decisioning system Intervention system Machine learning Supervised machine learning Unsupervised learning Reinforcement learning Data classification Conclusion References 5. Applying AI with Structured Data—Banking Introduction Structure Banking Credit scoring Credit bureaus Data in the credit bureaus Using the credit score from the bureau by the financial institutions Fraud detection systems Account takeover frauds Anomaly detection and the need for machine learning Example of anomaly dataset Anomaly detection with unsupervised Anomalies using a supervised method Offline validations Interventions Anti-Money Laundering (AML) AML with heuristics Problem with heuristics AML dataset Machine Learning for AML Conclusion References 6. Applying AI with Structured Data—Ecommerce Introduction Personalization Types of personalization Personalization types based on presentation Personalization types based on components Deep dive on content personalization Build an ML-based ranker Demand forecasting Conclusion References 7. Applying AI with Structured Data—On-Demand Deliveries Introduction On-demand hyperlocal deliveries AI use-cases Predicting ETA Predicting time components (t1) Predicting time component (t2) Predicting time component (t3) The metrics Surge pricing Supply-demand curves Solving for dynamic surge ML surge pricing Conclusion 8. AI in Natural Language Processing Introduction NLP overview Popular NLP use cases Searches Machine translations Reviews Social media listening Intelligent agents Content recommendations Automated insights NLP applications by verticals NLP in law firms NLP in e-commerce NLP in journalism NLP in customer service Algorithms and linguistics ML algorithms for NLP Supervised algorithms Term—document matrix with a bag of words Pre-trained embeddings Unsupervised algorithms Rule-based Distance-based Topic modeling Common NLP techniques Opinion mining Named Entity Recognition (NER) Information retrieval (finding the needle in the haystack) Crawling Indexing Retrieval Ranking Text summarization Intent mining Intent Mining in speech Dialog systems General purpose dialog systems Goal-oriented dialog systems NLP components State of the art A word of caution on sequence-sequence models Current challenges in NLP General context or “Knowledge of the world” Ambiguity in language Lexical ambiguity Syntax level ambiguity Referential ambiguity Sarcasm NLP for non-English Conclusion References 9. Bringing It All Together Introduction Where should you start the AI journey for your organization? Complexity The scale of data Actionability The team Data engineering Business intelligence (BI) Analytics Data science (DS) ML engineering The skillmap Building the data organization When AI is the backbone Special case of analytics services organization When AI is not the backbone The org structure Central data teams The pod approach The hybrid approach Common pitfalls Applying AI to everything Believing in the black box Tweaking it too much Delivery versus research Hesitation to experiment The success of data science solutions Conclusion Index
Donate to keep this site alive
1. Disable the AdBlock plugin. Otherwise, you may not get any links.
2. Solve the CAPTCHA.
3. Click download link.
4. Lead to download server to download.