Scribd
Upload
8 views1 page

Big Data Analytics Short Notes

The document provides an overview of key concepts in Big Data Analytics, including comparisons between NoSQL and NewSQL, the MapReduce framework, challenges of RDBMS and solutions offered by HDFS, types of learning in machine learning, and the role of graph databases in analytics. It outlines the characteristics and limitations of various data management systems and techniques. Additionally, it discusses dimensionality reduction and reinforcement learning in the context of big data.

Uploaded by

mansoorlashari2002
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
8 views1 page

Big Data Analytics Short Notes

The document provides an overview of key concepts in Big Data Analytics, including comparisons between NoSQL and NewSQL, the MapReduce framework, challenges of RDBMS and solutions offered by HDFS, types of learning in machine learning, and the role of graph databases in analytics. It outlines the characteristics and limitations of various data management systems and techniques. Additionally, it discusses dimensionality reduction and reinforcement learning in the context of big data.

Uploaded by

mansoorlashari2002
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

Big Data Analytics - Exam Short Notes

Q.01 (A) NoSQL vs NewSQL - NoSQL: Flexible schema, eventual consistency, API-based,
massive scalability. - NewSQL: SQL + ACID, relational schema, horizontal scaling. - NoSQL limits:
No ACID, hard joins, fragmented tooling → NewSQL created. Q.01 (B) MapReduce - Map: split
data, key/value output. - Reduce: aggregate results. - Pros: scalable, fault tolerant, works on any
data. - Cons: high latency, not for interactive queries. Q.02 RDBMS Challenges & HDFS
Solutions - Challenges: scaling limits, rigid schema, join overhead, storage bottlenecks. - HDFS:
horizontal scaling, replication, high throughput, stores any format. Q.03 (A) Learning Types -
Supervised: labels, prediction, e.g. spam filter. - Unsupervised: no labels, pattern discovery, e.g.
clustering. Q.03 (B) SVM - Finds max-margin hyperplane, uses kernels, good for high-dimensional
small data. Q.04 (A) RL in Big Data - Learns via reward feedback, handles sequential decisions,
uses Deep RL for big data. Q.04 (B) Dimensionality Reduction - Reduces features, removes
noise, speeds computation, visualizes data. Q.05 Graph DB & Analytics - Graph DB: nodes &
edges for relationships, efficient traversal. - Analytics types: centrality, communities, shortest paths,
components, motifs, embeddings.

You might also like