1. 732A54 Big Data Analytics 6hp

2. Teachers Examiner: Patrick Lambrix (B:2B:474)
Teachers
Examiner: Patrick Lambrix (B:2B:474)
Lectures: Patrick Lambrix,
Christoph Kessler,
Jose Pena,
Valentina Ivanova,
Johan Falkenjack
Labs: Zlatan Dragisic,
Valentina Ivanova
Director of studies: Patrick Lambrix

3. Course literature Articles (on web/handout) Lab descriptions (on web)
Course literature
Articles (on web/handout)
Lab descriptions (on web)

4.
Data and Data Storage

5. Data and Data Storage Database / Data source
Data and Data Storage
Database / Data source
One (of several) ways to store data in electronic format
Used in everyday life: bank, hotel reservations, library search, shopping

6. Databases / Data sourcces
Databases / Data sourcces
Database management system (DBMS): a collection of programs to create and maintain a database
Database system = database + DBMS

7. Databases / Data sources
Databases / Data sources
Information
Model
Queries
Answer
Database
system
Physical
database
management
Processing of
queries/updates
Access to stored data

8. What information is stored?
What information is stored?
Model the information
- Entity-Relationship model (ER)
- Unified Modeling Language (UML)

9. What information is stored? - ER
What information is stored? - ER
entities and attributes
entity types
key attributes
relationships
cardinality constraints
EER: sub-types

10.
1 tgctacccgc gcccgggctt ctggggtgtt ccccaaccac ggcccagccc tgccacaccc
61 cccgcccccg gcctccgcag ctcggcatgg gcgcgggggt gctcgtcctg ggcgcctccg
121 agcccggtaa cctgtcgtcg gccgcaccgc tccccgacgg cgcggccacc gcggcgcggc
181 tgctggtgcc cgcgtcgccg cccgcctcgt tgctgcctcc cgccagcgaa agccccgagc
241 cgctgtctca gcagtggaca gcgggcatgg gtctgctgat ggcgctcatc gtgctgctca
301 tcgtggcggg caatgtgctg gtgatcgtgg ccatcgccaa gacgccgcgg ctgcagacgc
361 tcaccaacct cttcatcatg tccctggcca gcgccgacct ggtcatgggg ctgctggtgg
421 tgccgttcgg ggccaccatc gtggtgtggg gccgctggga gtacggctcc ttcttctgcg
481 agctgtggac ctcagtggac gtgctgtgcg tgacggccag catcgagacc ctgtgtgtca
541 ttgccctgga ccgctacctc gccatcacct cgcccttccg ctaccagagc ctgctgacgc
601 gcgcgcgggc gcggggcctc gtgtgcaccg tgtgggccat ctcggccctg gtgtccttcc
661 tgcccatcct catgcactgg tggcgggcgg agagcgacga ggcgcgccgc tgctacaacg
721 accccaagtg ctgcgacttc gtcaccaacc gggcctacgc catcgcctcg tccgtagtct
781 ccttctacgt gcccctgtgc atcatggcct tcgtgtacct gcgggtgttc cgcgaggccc
841 agaagcaggt gaagaagatc gacagctgcg agcgccgttt cctcggcggc ccagcgcggc
901 cgccctcgcc ctcgccctcg cccgtccccg cgcccgcgcc gccgcccgga cccccgcgcc
961 ccgccgccgc cgccgccacc gccccgctgg ccaacgggcg tgcgggtaag cggcggccct
1021 cgcgcctcgt ggccctacgc gagcagaagg cgctcaagac gctgggcatc atcatgggcg
1081 tcttcacgct ctgctggctg cccttcttcc tggccaacgt ggtgaaggcc ttccaccgcg
1141 agctggtgcc cgaccgcctc ttcgtcttct tcaactggct gggctacgcc aactcggcct
1201 tcaaccccat catctactgc cgcagccccg acttccgcaa ggccttccag ggactgctct
1261 gctgcgcgcg cagggctgcc cgccggcgcc acgcgaccca cggagaccgg ccgcgcgcct
1321 cgggctgtct ggcccggccc ggacccccgc catcgcccgg ggccgcctcg gacgacgacg
1381 acgacgatgt cgtcggggcc acgccgcccg cgcgcctgct ggagccctgg gccggctgca
1441 acggcggggc ggcggcggac agcgactcga gcctggacga gccgtgccgc cccggcttcg
1501 cctcggaatc caaggtgtag ggcccggcgc ggggcgcgga ctccgggcac ggcttcccag
1561 gggaacgagg agatctgtgt ttacttaaga ccgatagcag gtgaactcga agcccacaat
1621 cctcgtctga atcatccgag gcaaagagaa aagccacgga ccgttgcaca aaaaggaaag
1681 tttgggaagg gatgggagag tggcttgctg atgttccttg ttg
Hint

11. DEFINITION Homo sapiens adrenergic, beta-1-, receptor
DEFINITION Homo sapiens adrenergic, beta-1-, receptor
ACCESSION NM_000684
SOURCE ORGANISM human
REFERENCE 1
AUTHORS Frielle, Collins, Daniel, Caron, Lefkowitz,
Kobilka
TITLE Cloning of the cDNA for the human
beta 1-adrenergic receptor
REFERENCE 2
AUTHORS Frielle, Kobilka, Lefkowitz, Caron
TITLE Human beta 1- and beta 2-adrenergic
receptors: structurally and functionally
related receptors derived from distinct
genes
PUT THIS SLIDE ALSO ON OH
We wil model this data using the different approaches to give a feel for the issues in the different approaches, we focus on data model + querying.
OBS: the formatted file resources have their own tags
(first step in a chain of biochemical reactions that put cell in stressed state)
SOURCE of data:Genbank
This membrane-bound protein is responsible for detecting adrenalin (ephineprine).
- adrenalin binds to the beta-receptor and induces a conformation change, which lets an associated membrane-bound G-protein lose a alpha sub-unit.
- alpha-unit changes a GDP toa GTP and acivates a neighboring membrane-bound adenylatcylase.
- the adenylatcylase synthesizes (converts ATP to) a lot of cAMP and then shuts down.
- the cAMP level in the cell activates a cAMP-dependent proteinkinase (cAPK) that mediates the signal to cellresponse through phosphorylation. (This phosphorylation puts the cell in a ’stressed state’.)

12. Entity-relationship protein-id source PROTEIN accession definition m
Entity-relationship
protein-id
source
PROTEIN
accession
definition m
Reference n
title
ARTICLE
article-id
author

13. Databases / Data sources
Databases / Data sources
Information
Model
Queries
Answer
Database
system
Physical
database
management
Processing of
queries/updates
Access to stored data

14.
How is the information stored? (high level) How is the information accessed? (user level)
structure
precision
Text (IR)
Semi-structured data
Data models (DB)
Rules + Facts (KB)

15. IR - formal characterization
IR - formal characterization
Information retrieval model: (D,Q,F,R)
D is a set of document representations
Q is a set of queries
F is a framework for modeling document representations, queries and their relationships
R associates a real number to document-query-pairs (ranking)

16. IR - Boolean model cloning adrenergic receptor Doc1 Doc2 (0 1 0) yes
IR - Boolean model
cloning
adrenergic
receptor
Doc1
Doc2
(1 1 0)
(0 1 0)
yes no
-->
Q1: cloning and (adrenergic or receptor)
--> (1 1 0) or (1 1 1) or (0 1 1) Result: Doc1
CONCEPTUAL MODEL
boolean model:
OBS: DISJUNCTIVE NORMAL FORM
disjunction of conjunction of literals
Q1: cloning and (adrenergic or receptor)
(cloning and adrenergic) or (cloning and receptor)
(cloning and adrenergic and receptor) or (cloning and adrenergic and not receptor) or (cloning and receptor and not adrenergic)
(1 1 1) or (1 1 0) or (0 1 1)
Q2: cloning and not adrenergic
(cloning and not adrenergic and receptor) or (cloning and not adrenergic and not receptor)
(0 1 1) or (0 1 0)
Q2: cloning and not adrenergic
--> (0 1 0) or (0 1 1) Result: Doc2

17. IR - Vector model (simplified)
IR - Vector model (simplified)
Doc1 (1,1,0)
cloning
receptor
adrenergic
Doc2 (0,1,0)
Q (1,1,1)
CONCEPTUAL MODEL
vector space model:
- define n-dimensional space with n = number of words
- represent each document as a vector
for each word, if word occurs in document, then 1 for coordinate corresponding to the word, else 0
- represent query in similar way
- angle between vectors is an indication of similarity
cos alpha = (d. q) / |d| |q|
OBS: |q| has no influence on the ranking
|d| gives a normalization
(length of d depends on the number of words in the document)
In this example: simplified model
Normally, the coordinates are weights representing importance and selectiveness of words in document or query.
sim(d,q) = d . q
|d| x |q|

18. Semi-structured data Protein DB ”Homo sapiens adrenergic, human
Semi-structured data
DEFINITION
SOURCE
human
”Homo sapiens adrenergic,
beta-1-, receptor”
AUTHOR
Frielle
Collins
Daniel
Caron
Lefkowitz
Kobilka
REFERENCE
ACCESSION
NM_000684
TITLE
”Cloning of …”
”Human beta-1 …”
PROTEIN
Protein DB

19. Semi-structured data - Queries
Semi-structured data - Queries
select source
from PROTEINDB.protein P
where P.accession = ”NM_000684”;

20. Relational databases PROTEIN ACCESSION SOURCE DEFINITION
Relational databases
PROTEIN
ACCESSION
SOURCE
DEFINITION
Homo sapiens adrenergic,
beta-1-, receptor
NM_000684
human
PROTEIN-ID 1
REFERENCE
ARTICLE-ID 2
ARTICLE-AUTHOR
Human beta 1- and beta 2-adrenergic receptors: structurally and functionally related receptors derived from distinct
genes
ARTICLE-ID
TITLE
Cloning of the cDNA for the human
beta 1-adrenergic receptor
ARTICLE-TITLE 1 2
ARTICLE-ID
AUTHOR 1 2
Frielle
Collins
Daniel
Caron
Lefkowitz
Kobilka

21. Relational databases - SQL
Relational databases - SQL
select source
from protein
where accession = NM_000684;
PROTEIN
ACCESSION
SOURCE
DEFINITION
Homo sapiens adrenergic,
beta-1-, receptor
NM_000684
human
PROTEIN-ID 1

22. Evolution of Database Technology
Data collection, database creation, IMS and network DBMS
1970s:
Relational data model, relational DBMS implementation
1980s:
Advanced data models (extended-relational, OO, deductive, etc.)
Application-oriented DBMS (spatial, temporal, multimedia, etc.)
1990s:
Data mining, data warehousing, multimedia databases, and Web databases
2000s
Stream data management and mining
Data mining and its applications
Web technology (XML, data integration) and global information systems
NoSQL databases

23. Knowledge bases (F) source(NM_000684, Human)
Knowledge bases
(F) source(NM_000684, Human)
(R) source(P?,Human) => source(P?,Mammal)
(R) source(P?,Mammal) => source(P?,Vertebrate)
Q: ?- source(NM_000684, Vertebrate)
A: yes
Q: ?- source(x?, Mammal)
A: x? = NM_000684
F. The source organism of NM_ is Human.
If the source organism is Human then it is also mammal
If …..mammal then …vertebrate
Is vetebrate the source organism of NM_000684?
A: inferred using F + 2 rules
Retrieve all the entries that have Mammal as source organism.
A: inferred using F + 1 rule

24. Interested in more? 732A57 Database Technology (relational databases)
Interested in more?
732A57 Database Technology
(relational databases)
TDDD43 Advanced data models and databases
(IR, semi-structured data, DB, KB)
732A47 Text mining
(includes IR)

25. Analytics

26.
Analytics
Discovery, interpretation and communication of meaningful patterns in data

27. Analytics - IBM What is happening? Descriptive
Discovery and explanation
Why did it happen? Diagnostic
Reporting, analysis, content analytics
What could happen? Predictive
Predictive analytics and modeling
What action should I take? Prescriptive
Decision management
What did I learn, what is best?
Cognitive

28. Analytics - Oracle Classification Regression Clustering
Attribute importance
Anomaly detection
Feature extraction and creation
Market basket analysis

29. Why Analytics? The Explosive Growth of Data
Data collection and data availability
Automated data collection tools, database systems, Web, computerized society
Major sources of abundant data
Business: Web, e-commerce, transactions, stocks, …
Science: Remote sensing, bioinformatics, scientific simulation, …
Society and everyone: news, digital cameras, YouTube
We are drowning in data, but starving for knowledge!

30. Ex.: Market Analysis and Management
Where does the data come from?—Credit card transactions, loyalty cards, discount coupons, customer complaint calls, plus (public) lifestyle studies
Target marketing
Find clusters of “model” customers who share the same characteristics: interest, income level, spending habits, etc.
Determine customer purchasing patterns over time
Cross-market analysis—Find associations/co-relations between product sales, & predict based on such association
Customer profiling—What types of customers buy what products (clustering or classification)
Customer requirement analysis
Identify the best products for different groups of customers
Predict what factors will attract new customers
Provision of summary information
Multidimensional summary reports
Statistical summary information (data central tendency and variation)
Market baket analysis: computer and printer sell well together

31. Ex.: Fraud Detection & Mining Unusual Patterns
Approaches: Clustering & model construction for frauds, outlier analysis
Applications: Health care, retail, credit card service, telecomm.
Auto insurance: ring of collisions
Money laundering: suspicious monetary transactions
Medical insurance
Professional patients, ring of doctors, and ring of references
Unnecessary or correlated screening tests
Telecommunications: phone-call fraud
Phone call model: destination of the call, duration, time of day or week. Analyze patterns that deviate from an expected norm
Anti-terrorism

32. Knowledge Discovery (KDD) Process
Pattern evaluation and presentation
Data Mining
Task-relevant Data
Selection and transformation
Data Warehouse
Data Cleaning
Preprocessing steps:
Data cleaning: remove noise and inconsistent data
Data integration: multiple sources can be combined
Data selection: data relevant to the analysis task is retrieved from the database
Data transformation: data transformed into forms appropriate for the mining (e.g. Summarizing, aggregating)
Data mining: extract data patterns using ’intelligent’ methods
Pattern evaluation: identify truly interesting patterns, interestingness measures
Knowledge presentation: present knowledge to user – visualization and knowledge representation
Data Integration
Databases

33. Data Mining: Classification Schemes
General functionality
Descriptive data mining
Predictive data mining
Descriptive: characterize the general properties of the data
Predicitive: perform inference on the current data to make predictions
Data to be mined
Relational, data warehouse, transactional, stream, object-oriented/relational, active, spatial,
time-series, text, multi-media, heterogeneous, legacy, WWW
Knowledge to be mined
Characterization, discrimination, association, classification, clustering, trend/deviation, outlier analysis, etc.
Multiple/integrated functions and mining at multiple levels
Techniques utilized
Database-oriented, data warehouse (OLAP), machine learning, statistics, visualization, etc.
Applications adapted
Retail, telecommunication, banking, fraud analysis, bio-data mining, stock market analysis,
text mining, Web mining, etc.

34. Data Mining – what kinds of patterns?
Concept/class description:
Characterization: summarizing the data of the class under study in general terms
E.g. Characteristics of customers spending more than sek per year
Discrimination: comparing target class with other (contrasting) classes
E.g. Compare the characteristics of products that had a sales increase to products that had a sales decrease last year

35. Data Mining – what kinds of patterns?
Frequent patterns, association, correlations
Frequent itemset
Frequent sequential pattern
Frequent structured pattern
E.g. buy(X, “Diaper”)  buy(X, “Beer”) [support=0.5%, confidence=75%]
confidence: if X buys a diaper, then there is 75% chance that X buys beer
support: of all transactions under consideration 0.5% showed that diaper and
beer were bought together
E.g. Age(X, ”20..29”) and income(X, ”20k..29k”)  buys(X, ”cd-player”) [support=2%, confidence=60%]
Frequent itemset: items frequently appearing together in transactional data (e.g. bread+milk; computer+printer)
Frequent sequential pattern: sequence patterns (e.g first buy computer, then camera, then memory card)
Frequent structured pattern: graphs, trees, lattices

36. Data Mining – what kinds of patterns?
Classification and prediction
Construct models (functions) that describe and distinguish classes or concepts for future prediction.
The derived model is based on analyzing training data – data whose class labels are known.
E.g., classify countries based on (climate), or classify cars based on (gas mileage)
Predict some unknown or missing numerical values
Decision trees, neural nets

37. Data Mining – what kinds of patterns?
Cluster analysis
Class label is unknown: Group data to form new classes, e.g., cluster customers to find target groups for marketing
Maximizing intra-class similarity & minimizing interclass similarity
Outlier analysis
Outlier: Data object that does not comply with the general behavior of the data
Noise or exception? Useful in fraud detection, rare events analysis
Trend and evolution analysis
Trend and deviation

38. Interested in more? 732A95 Introduction to machine learning
Interested in more?
732A95 Introduction to machine learning
732A61 Data mining – clustering and association analysis

39.
Big Data

40.
Big Data
So large data that it becomes difficult to process it using a ’traditional’ system
Eg. 100 MB you cannot send
100 GB image you cannot view
100 TB video you cannot edit
Kilo (3), mega (6), giga (9), tera (12), peta (15), exa (18), zetta (21), yotta (24)

41.
Big Data – 3Vs
Volume
size of the data

42. Volume - examples Facebook processes 500 TB per day
Walmart handles 1 million customer transaction per hour
Airbus generates 640 TB in one fligth (10 TB per 30 minutes)
72 hours of video uploaded to youtube every minute
SMS, , internet, social media

43. https://y2socialcomputing.files.wordpress.com/2012/06/
social-media-visual-last-blog-post-what-happens-in-an-internet-minute-infographic.jpg

44. Big Data – 3Vs Volume Variety size of the data
Big Data – 3Vs
Volume
size of the data
Variety
type and nature of the data
text, semi-structured data, databases, knowledge bases
10% of the data is structured
Challenge: to extract meaningful information

45. Linked open data
Linking Open Data cloud diagram 2014, by Max Schmachtenberg, Christian Bizer, Anja Jentzsch and Richard Cyganiak.

46. Linked open data of US government
Format   (# Datasets)
  HTML (27005)
  XML  (24077)
  PDF  (19628)
  CSV  (10058)
  JSON (8948)
  RDF (6153)
  JPG (5419)
  WMS (5019)
  Excel (3389)
  WFS (2781)

47. Big Data – 3Vs Volume Variety Velocity size of the data
Big Data – 3Vs
Volume
size of the data
Variety
type and nature of the data
Velocity
speed of generation and processing of data

48. Velocity - examples
Traffic data
Financial market
Social networks

50. Big Data – other Vs Variability Veracity Value …
Big Data – other Vs
Variability
inconsistency of the data
Veracity
quality of the data
Value
useful analysis results …

51.
Applications

52. BDA system architecture
Specialized services
for domain A
Specialized services
for domain B
Big Data Services Layer
Knowledge Management Layer
Data storage and management layer: Development of a general software environment for computational sciences in big data domains. This software environment must exhibit flexibility in terms of the data storage and data management solutions so as to guarantee its generality. The suitability of such solutions is highly dependent on the application area. Therefore, for each application area, an area-specific definition of properties that make such solutions suitable requires a close collaboration between domain experts and data management systems experts.
Knowledge management layer: Integration of various semantic technologies such as ontologies and graphical models. Graphical models can be used to improve and merge existing ontologies which, in turn, can be used to assist in the learning of the final graphical models which, in turn, can be used to produce human-understandable reasoning. Due to these dependencies, the benefit of such integration can be maximized only by a coordinated effort of partners with expertise in these different types of semantic technologies.
Big data services layer: Development of parallel algorithms for learning graphical models from big data, and reasoning with big models. Our project environment provides a unique opportunity to leverage knowledge from domain experts in machine learning, computer architectures, software design, and performance of distributed systems within the same environment. This is necessary to achieve the best results.
Data Storage and Management Layer

53. BigMecs
Specialized services for materials design: Deployment of the knowledge system built in the project to design materials with the desired features efficiently. In addition to important contributions to material sciences, we expect that the integration of this service and the other two specialized services into our complete system framework will help demonstrate the power and flexibility of our system. Without a cross-disciplinary research team this would not be possible.
Specialized services for environmental sciences: Deployment of the knowledge system built in the project to simulate and analyze models of freshwater flow and nutrients for large parts of the world, improving existing techniques in terms of processing speed, quality of the results, and interpretability of the results.
Specialized services for computational criminology: Deployment of the knowledge system built in the project to extract actionable information from large information streams in real-time, as well as to perform post processing and reasoning to help the law enforcement in their duties.

54. BDA system architecture
Large amounts of data, distributed environment
Unstructured and semi-structured data
Not necessarily a schema
Heterogeneous
Streams
Varying quality
Data storage and management layer: Development of a general software environment for computational sciences in big data domains. This software environment must exhibit flexibility in terms of the data storage and data management solutions so as to guarantee its generality. The suitability of such solutions is highly dependent on the application area. Therefore, for each application area, an area-specific definition of properties that make such solutions suitable requires a close collaboration between domain experts and data management systems experts.
Knowledge management layer: Integration of various semantic technologies such as ontologies and graphical models. Graphical models can be used to improve and merge existing ontologies which, in turn, can be used to assist in the learning of the final graphical models which, in turn, can be used to produce human-understandable reasoning. Due to these dependencies, the benefit of such integration can be maximized only by a coordinated effort of partners with expertise in these different types of semantic technologies.
Big data services layer: Development of parallel algorithms for learning graphical models from big data, and reasoning with big models. Our project environment provides a unique opportunity to leverage knowledge from domain experts in machine learning, computer architectures, software design, and performance of distributed systems within the same environment. This is necessary to achieve the best results.
Data Storage and Management Layer

55. Data Storage and management – this course
NoSQL databases
OLTP vs OLAP
Horizontal scalability
Consistency, availability, partition tolerance
Data management
Hadoop
Data management systems
OLTP – traditional applications we are used to think about when talking about databases: University Database – students, teacher, courses; airline ticket reservations; credit card processing
Many concurrent users; high availability and fast response times for all users; many reads, writes and updates
OLAP - - Data warehouses for more complex analysis
Preloaded, mainly reads, optimized for read operations
Online analytical processing (OLAP) OLAP is characterized by a relatively low volume of transactions. Queries are often very complex and involve aggregations. For OLAP systems, response time is an effectiveness measure. OLAP applications are widely used by Data Mining techniques. OLAP databases store aggregated, historical data in multi-dimensional schemas (usually star schemas). OLAP systems typically have data latency of a few hours. OLAP approach is used to analyze multidimensional data from multiple sources and perspectives. The three basic operations in OLAP are : Roll-up (Consolidation), Drill-down and Slicing & Dicing.[3]
Online transaction processing (OLTP) OLTP is characterized by a large number of short on-line transactions (INSERT, UPDATE, DELETE). OLTP systems emphasize very fast query processing and maintaining data integrity in multi-access environments. For OLTP systems, effectiveness is measured by the number of transactions per second. OLTP databases contain detailed and current data. The schema used to store transactional databases is the entity model (usually 3NF).[4] Normalization is the norm for data modeling techniques in this system.

56. BDA system architecture
Semantic technologies
Integration
Knowledge acquisition
Data storage and management layer: Development of a general software environment for computational sciences in big data domains. This software environment must exhibit flexibility in terms of the data storage and data management solutions so as to guarantee its generality. The suitability of such solutions is highly dependent on the application area. Therefore, for each application area, an area-specific definition of properties that make such solutions suitable requires a close collaboration between domain experts and data management systems experts.
Knowledge management layer: Integration of various semantic technologies such as ontologies and graphical models. Graphical models can be used to improve and merge existing ontologies which, in turn, can be used to assist in the learning of the final graphical models which, in turn, can be used to produce human-understandable reasoning. Due to these dependencies, the benefit of such integration can be maximized only by a coordinated effort of partners with expertise in these different types of semantic technologies.
Big data services layer: Development of parallel algorithms for learning graphical models from big data, and reasoning with big models. Our project environment provides a unique opportunity to leverage knowledge from domain experts in machine learning, computer architectures, software design, and performance of distributed systems within the same environment. This is necessary to achieve the best results.
Knowledge Management Layer

57. Knowledge management – this course
Not a focus topic in this course
For semantic and integration approaches see TDDD43

58. BDA system architecture
Analytics services for Big Data
Data storage and management layer: Development of a general software environment for computational sciences in big data domains. This software environment must exhibit flexibility in terms of the data storage and data management solutions so as to guarantee its generality. The suitability of such solutions is highly dependent on the application area. Therefore, for each application area, an area-specific definition of properties that make such solutions suitable requires a close collaboration between domain experts and data management systems experts.
Knowledge management layer: Integration of various semantic technologies such as ontologies and graphical models. Graphical models can be used to improve and merge existing ontologies which, in turn, can be used to assist in the learning of the final graphical models which, in turn, can be used to produce human-understandable reasoning. Due to these dependencies, the benefit of such integration can be maximized only by a coordinated effort of partners with expertise in these different types of semantic technologies.
Big data services layer: Development of parallel algorithms for learning graphical models from big data, and reasoning with big models. Our project environment provides a unique opportunity to leverage knowledge from domain experts in machine learning, computer architectures, software design, and performance of distributed systems within the same environment. This is necessary to achieve the best results.
Big Data Services Layer

59. Big Data Services – this course
Big data versions of analytics/data mining algorithms
OLTP – traditional applications we are used to think about when talking about databases: University Database – students, teacher, courses; airline ticket reservations; credit card processing
Many concurrent users; high availability and fast response times for all users; many reads, writes and updates
OLAP - - Data warehouses for more complex analysis
Preloaded, mainly reads, optimized for read operations
Online analytical processing (OLAP) OLAP is characterized by a relatively low volume of transactions. Queries are often very complex and involve aggregations. For OLAP systems, response time is an effectiveness measure. OLAP applications are widely used by Data Mining techniques. OLAP databases store aggregated, historical data in multi-dimensional schemas (usually star schemas). OLAP systems typically have data latency of a few hours. OLAP approach is used to analyze multidimensional data from multiple sources and perspectives. The three basic operations in OLAP are : Roll-up (Consolidation), Drill-down and Slicing & Dicing.[3]
Online transaction processing (OLTP) OLTP is characterized by a large number of short on-line transactions (INSERT, UPDATE, DELETE). OLTP systems emphasize very fast query processing and maintaining data integrity in multi-access environments. For OLTP systems, effectiveness is measured by the number of transactions per second. OLTP databases contain detailed and current data. The schema used to store transactional databases is the entity model (usually 3NF).[4] Normalization is the norm for data modeling techniques in this system.

60. Databases
Parallel
programming
Machine
learning

61. 2016: course for SDM year 1 and year 2
2016: course for SDM year 1 and year 2
 some review/introduction lectures

62. Course overview Review Databases (lectures + labs)
Course overview
Review
Databases (lectures + labs)
Python (lectures + labs)
Databases for Big Data (lectures + lab)
Parallel algorithms for processing Big Data (lectures + lab + exercise session)
Machine Learning for Big Data (lectures + lab)
Visit to National Supercomputer Centre

63. Info Results reported in connection to exams
Info
Results reported in connection to exams
Info about handing in labs on web; strong recommendation to hand in as soon as possible
Sign up for labs via web (in pairs)

64. Info Relational database labs require special database account
Info
Relational database labs require special database account
 make sure you are registered for the course
BDA labs require special access to NSC resources
 fill out forms

65.
Info
Lab deadlines:
Final deadlines in connection to the exams; no reporting between exams
HARD DEADLINE: March exam
(No guarantee NSC resources available after April.)

66.
Examination
Written exam
Labs

67. What if I already took …? What if I also take…?
TDDD37/732A57 Database technology
RDB labs 1-2 in one of the courses, results registered for both
732A47 Text mining
Python labs in one of the courses, results registered for both

68. Changes w.r.t. last year
New course

69. My own interest and research
My own interest and research
Modeling of data
Ontologies
Ontology engineering
Ontology alignment
(Winner Anatomy track OAEI 2008 /
Organizer OAEI tracks since 2013)
Ontology debugging
(Founder and organizer WoDOOM/CoDeS since 2012)
Ontologies and databases for Big Data
Former work: knowledge representation, data integration, knowledge-based information retrieval, object-centered databases