Neo4j Introduction Workshop for Partners

Neo4j
Introduction Workshop
for Partners

Agenda
1. What is a graph database?
2. The Neo4j differences
3. Graph database concepts
4. GDS, GenAI, 7 graphs in an enterprise
Neo4j Inc. All rights reserved 2023
2
M. Arif Wicaksana
Sales Engineer
arif@neo4j.com

What is
a graph database?
3

What is Graph?
4
This is a Graph
Mathematical and Data structures used
to model pairwise relations between
objects
This is a Graph Database
A database technology that stores, queries
and computes data in the shape of graphs
Not This Graph
Creator of the Labeled Property Graph

Anything can be a graph
A water molecule
H
O
H
The metro
Social media
5

Relationships connect
nodes and represent
actions
Relationships can have
properties
(key:value pairs)
Relationships
are directional
Nodes represent
objects
Nodes can have
properties
The power of the
graph model

Graph is fast
Response
Time
Better
Performance
Number of Hops
Relational DB
Graph DB

Employees
Network & Security
Suppliers
Product
Customers
Finance
Process
Graph is insightful
What’s unusual?
What’s important?
What’s next?

The Neo4j Differences
10

11
300
$1B+ Enterprise
customers
$500M
In funding
170+
Global Partner
Ecosystem
250K
Community of
developers and
data pros
100M+
Downloads
The ﬁrst-ever
graph
database
Creator of the
market
category
Continued
market leader

is enterprise-grade
No limit on
the database
size, only
depending on
your
infrastructure
ACID and
OLAP,
Capture
Data
Change,
Backups and
Restore…
TB Graphs
with
sharding
Integrations
with
On Premise
or any
cloud:
any
workload SSO, SSL
Encryption
Complete
RBAC
framework
Ops
Manager
Neo4j
Databases
Management
Tool
Graph
algorithms
Uniﬁed DB for analytical
& operational workloads

is enterprise-grade
Up to
Vertical
scaling
Guaranteed
Uptime
SLA
TB Graphs
with
sharding
Integrations
with
Any cloud.
Any
workload
Encrypted
data
SOC2
Type 2
HIPAA
compliant
Graph
algorithms
Uniﬁed DB for analytical
& operational workloads

Neo4j Graph Database Capabilities
14
Engineering Expertise
>1000 person-years investment
First mover advantage
Maturity, Most enterprise deployments
Largest graph community
Growing at 80%+ annually
Hybrid
Workloads
Native Graph
Architecture
Powers Graph
Data Science
Rich
Toolset
Enterprise
Trust &
Security
Runs
Anywhere

Graph Applications
(System of Record Applications)
Application Stack
Real-time Messaging & Processing
(Clickstream, IOT, CDC etc)
Developer
Intelligent apps
Data Analyst
Query and analyze
Biz Analyst
Visual Analytics
Data Scientist
Algorithms and features
Data engineer
Get clean, useful data
ML Engineer
ML Ops
Data Platform
(Data Platforms-Snowflake, BigQuery,
Databricks, Elastic)
AI/ML Ops
BI Platforms
Graph Analytics & Visualizations
(Real-time applications, visualizations
and algorithms)
Powered by
How does Neo4j fit in?
Modern Enterprise Data Platform
15
Neo4j’s knowledge
graph platform
augments existing
enterprise data
investments by
providing unique and
valuable business
insights through
efficient, large-scale
analyses of
relationships hidden
in your data

Graph Database
Concepts
16

© 2023 Neo4j, Inc. All rights reserved.
NATIVE GRAPH ARCHITECTURE: FAST, FLEXIBLE, SCALABLE
Native Graph Database Native Graph Storage
✔ No model mismatch
✔ Data integrity with ACID
✔ Flexible schema
Native Graph Processing
✔ 1000x faster than relational
✔ Easy-to-use, native graph query language
Scalable sharded Clusters
✔ Autonomous Clustering
✔ Unlimited high throughput
✔ Elastic scale-out to 100s
of machines across clusters
✔ Federated queries of scaled out shards
✔ Instant composite database
At write time
Data is connected
as it is stored
At read time
Lightning-fast retrieval of data and
relationships via pointer chasing
17

Graph Enables:
1. More sophisticated queries (you can
ask questions that were previously
impossible)
2. Real-time applications and insight (get
your complex queries answered in
milliseconds)
3. Better Data Science (running graph
algorithms on contextualized,
connected data leads to better AI &
ML outcomes)
4. Future-proof applications (it takes
minutes, not months to change your
schema)
Performance Leads to Signiﬁcantly More Value
18

SQL Joins–using Pkey Indexes
19
Orders
OrderItems
OrderItems PK index
{Order_num, Item_num}
<m
<f5
>c11
<d112
Order_num=e1234567
<e4
>e120
Products
Products PK index
{product_num}
Manufacturer
Retailer
6 IO’s traversing the index
(plus a few if leaf is >1 page)
(do 6 times – 1 for each row)
1 outer row + 7 index + 6 inner rows + 6(6 index + 1 row) +
6(5 index + 1 row) + 6(5 index + 1) row = 128 IO’s

20
Orders
OrderItems
<m
<f5
>c11
<d112
<e4
>e120
Products
-[:ORDERED_PRODUCT]🡪
Manufacturer
Retailer
1 outer node + 6 inner nodes + 6(p nodes) + 6(m nodes)
+ 6 (r nodes) = 25 IO’s
Neo4j Graph Traversals (Index Free Adjacency)
1 direct pointer
1 direct pointer
Order_num=e1234567
6 direct pointers 1 direct pointer
-[:HAS_ITEMS]🡪 -[:MADE_BY]🡪
-[:SOLD_BY]🡪

Only 20 indexes were required to deliver
millisecond response time at scale.
In contrast to Neo4j’s graph model, a relational
database approach would have an
entity-relationship diagram with over 150
tables and 300 to 400 indexes (not including
primary keys) for the same questions and
dataset. The relational model would also
require hundreds of indexes to deliver much
slower query-response times measured in
minutes or hours.
Reference:
https://neo4j.com/blog/combating-money-laundering-sprints-with-money-queries/

Labelled property graph
22

Toy
Story
Nodes represent things

Toy
Story
Movie

Toy
Story
Movie
Nodes can be identiﬁed by
one or more labels

Toy
Story
Movie
Animated
one or more labels

Toy
Story
Movie
Animated
one or more labels
title: Toy Story
released: 1995

Toy
Story
Movie
Animated
one or more labels
title: Toy Story
released: 1995
Nodes can hold properties
as key/value pairs

Tom
Hanks
Actor
Toy
Story
Movie

Toy
Story
Movie
Tom
Hanks
Actor

Toy
Story
Movie
Tom
Hanks
Actor
Relationships connect two nodes

Toy
Story
Movie
Tom
Hanks
Actor
ACTED_IN

Toy
Story
Movie
Tom
Hanks
Actor
ACTED_IN
Relationships
have a type

Toy
Story
Movie
Tom
Hanks
Actor
ACTED_IN
Relationships
have a type
Relationships
have a direction

Toy
Story
Movie
Tom
Hanks
Actor
ACTED_IN
roles: Woody
Relationships
have a type
Relationships
have a direction

Toy
Story
Movie
Tom
Hanks
Actor
ACTED_IN
roles: Woody
Relationships
have a type
Relationships can also
hold properties as
key/value pairs
Relationships
have a direction

Toy
Story
Movie
ACTED_IN
Tom
Hanks
Actor

42
GQL ISO Standard Published -April 2024
● Establishes GQL as the ofﬁcial,
internationally accepted
graph query language
● Founded on Neo4j’s Cypher language
● First ISO data query language
certiﬁcation since SQL in 1987
● Drives enterprise graph adoption,
vendor interoperability, graph R&D,
foundational skilling and technical
ubiquity
https://neo4j.com/press-releases/gql-standard/

Cypher is a declarative language that allows you
to identify patterns in your data using an
ASCII-art style syntax consisting of brackets,
dashes and arrows.
Movie
ACTED_IN
Person
(p:Person)-[r:ACTED_IN]->(m:Movie)

dashes and arrows.

dashes and arrows.
Movie
ACTED_IN
Person

Creating a Neo4j
Sandbox
55

56
https://sandbox.neo4j.com/
Free, Easy and Fast!
New user ? Existing user ?

57
Pick your project

58
Many options…
Some are Blanks for your
own dataset
Some have Graph Data
Science (GDS) activated
Some have GraphQL
activated

59
Pick the Top Right: Graph Data Science
You should see your
project and have a ﬁle
with credentials
downloaded

60
All available in one click
See all the details and
the access to the Neo4j
tools
See the links for
Documentation and
GraphAcademy

61
Let’s try it…

Reading from Neo4j
// Find a pattern in the database
MATCH (p:Person)-[r:ACTED_IN]->(m:Movie)

Reading from Neo4j
// Filter on a node property
WHERE m.title = "Toy Story"

WHERE m.title = "Toy Story"
// Choose what to return
RETURN p.name AS actor, r.roles AS roles
Reading from Neo4j

// Find a pattern in the database SQL Equivalent
MATCH (p:Person)-[r:ACTED_IN]->(m:Movie) FROM/JOIN
WHERE m.title = "Toy Story" WHERE
// Choose what to return
RETURN p.name AS actor, r.roles AS roles SELECT
Reading from Neo4j

MATCH p=(n:Actor)-[:ACTED_IN]-(:Movie)-[:LOCATED_AT]-(:Country)
WHERE n.name=’’
RETURN p

Have a go!
1. Find a Movie you
like.
2. Use the ACTED_IN
relationship to ﬁnd
who acted in the
movie.
3. Use the DIRECTED
relationship to ﬁnd
who directed the
movie
MATCH (m:Movie) RETURN m.title
MATCH (m:Movie {title: “???”})-[:??????]->(p:Person)
RETURN p.name
MATCH (m:Movie {title: “???”})-[:??????]->(p:Person)
RETURN p.name

Writing to Neo4j
// Find or create using MERGE
MERGE (p:Person {name: "Tom Hanks"})

Writing to Neo4j
MERGE (m:Movie {title: "Toy Story"})

Writing to Neo4j
MERGE (p)-[r:ACTED_IN]->(m)

Writing to Neo4j
MERGE (p)-[r:ACTED_IN]->(m)
SET r.roles = "Woody"

1. Create a Person
node for yourself
2. Give yourself a role
in a movie by
creating an
ACTED_IN
relationship
3. Find your role in the
movie
Have a go!
MERGE (me:Person{name: “??????”})
MATCH (me:Person{name: “??????”})
MATCH (m:Movie{title: “??????”})
MERGE (me)-[r:??????]->(m)
SET r.role = “???????”
MATCH (p:Person{name:”??????”})-[r:ACTED_IN]->(????)
RETURN ?.?????

Nouns, Verbs and Adjectives
78
AS A
Customer
I WOULD LIKE TO
Order a new Laptop and accessories
SO THAT I CAN
Build a website using a graph database

79
AS A
Customer
I WOULD LIKE TO
SO THAT I CAN
Product.

80
AS A
Customer
I WOULD LIKE TO
SO THAT I CAN
Product.

81
AS A
Customer
I WOULD LIKE TO
SO THAT I CAN
Product.

Intermediate Nodes
82

Intermediate Nodes
83
Knowing this fact might be
enough for your use case

Intermediate Nodes
84

Intermediate Nodes
85
The order
information can
now be linked
to multiple
nodes

Relational
to Graph

Relational
to Graph
Table names become
labels, individual records
become nodes

Relational
to Graph
Foreign keys become
relationships
(:Supplier)-[:SUPPLIES]->(:Product)
(:Customer)-[:PURCHASED]->(:Order)
(:Shipper)-[:SHIPS]->(:Order)

The Northwind
Database
Many-to-Many
Relationships

The Northwind
Database
Many-to-Many
Relationships
��

The Northwind
Database
Many-to-Many
Relationships
Order
C
O
N
T
A
I
N
S
PRODUCT

The Northwind
Database
Many-to-Many
Relationships
Order
C
O
N
T
A
I
N
S
PRODUCT
u
n
i
t
P
r
i
c
e
:
n
u
m
b
e
r
q
u
a
n
t
i
t
y
:
n
u
m
b
e
r
d
i
s
c
o
u
n
t
:
f
l
o
a
t

RDBMS to Graph in summary…
97
Relational
Rows
Joins
Table Names
Columns
Graph
Nodes
Relationships
Labels
Properties

Property, Relationship or Node - it depends…

{
"customer": {
"CustomerID": "ALFKI",
"CompanyName": "Alfreds Futterkiste",
"ContactName": "Maria Anders",
"ContactTitle": "Sales Representative",
"Phone": "030-0074321",
"Fax": "030-0076545",
"email": "maria.anders@alfreds.de"
},
"address": {
"Address": "Obere Str. 57",
"City": "Berlin",
"Region": null,
"PostalCode": "12209",
"Country": "Germany"
},
"order": {
"OrderID": 10248,
"EmployeeID": 5,
"OrderDate": "1996-07-04T00:00:00",
"RequiredDate": "1996-08-01T00:00:00",
"ShippedDate": "1996-07-16T00:00:00",
"ShipVia": 3,
"Freight": 32.38,
"ShipName": "Alfreds Futterkiste",
"ShipAddress": {
"City": "Berlin",
"Region": null,
}
}
}

● Many databases will duplicate data for
read-time performance
● Address is read-only? Store as a property
● Are billing and shipping addresses saved as
part of the checkout process? Store in a Node
● Are offers served to ﬁrst-time customers that
are limited by address? Store in a Node
● Is the graph used to optimise delivery routes?
Store as a Node
{
"customer": {
"CompanyName": "Alfreds Futterkiste",
"ContactName": "Maria Anders",
"ContactTitle": "Sales Representative",
"Phone": "030-0074321",
"Fax": "030-0076545",
"email": "maria.anders@alfreds.de"
},
"address": {
"City": "Berlin",
"Region": null,
},
"order": {
"OrderID": 10248,
"EmployeeID": 5,
"OrderDate": "1996-07-04T00:00:00",
"RequiredDate": "1996-08-01T00:00:00",
"ShippedDate": "1996-07-16T00:00:00",
"ShipVia": 3,
"Freight": 32.38,
"ShipName": "Alfreds Futterkiste",
"ShipAddress": {
"City": "Berlin",
"Region": null,
}
}
}

Introduction to
Graph Data Science (GDS)

Graph Data Science: make sense of data relationships
102
Machine Learning Pipeline
Pl
ay
s
Lives_in
In_sport
Likes
F
a
n
_
o
f
Plays_for
K
n
o
w
s
Knows
Knows
K
n
o
w
s
Explore the hidden patterns and features in your data
What’s important? What’s unusual? What’s next?

Graph and Data Science
103
Knowledge Graphs
Graph Algorithms
Graph Native
Machine Learning
Find the patterns you’re
looking for in connected data
Use machine learning
techniques to identify
associations, anomalies, and
trends.
Learn the features in your
graph that you don’t even
know are important yet.
Train in-graph supervised ML
models to predict links,
labels, and missing data.

105
Example: Graph Algorithms for Detecting Fraud
Graph algorithms enable reasoning
about network structure

106
Connected components
identify disjointed group
sharing identiﬁers

107
PageRank to measure inﬂuence
and transaction volumes

108
Louvain to identify
communities that frequently
interact

109
Jaccard to measure account
similarity

75+ Graph Data Science Techniques in Neo4j
110
Pathfinding &
Search
• Shortest Path
• Single-Source Shortest Path
• All Pairs Shortest Path
• A* Shortest Path
• Yen’s K Shortest Path
• Minimum Weight Spanning Tree
• K-Spanning Tree (MST)
• Random Walk
• Breadth & Depth First Search
Centrality &
Importance
• Degree Centrality
• Closeness Centrality
• Harmonic Centrality
• Betweenness Centrality & Approx.
• PageRank
• Personalized PageRank
• ArticleRank
• Eigenvector Centrality
• Hyperlink Induced Topic Search (HITS)
• Influence Maximization (Greedy, CELF)
Community
Detection
• Triangle Count
• Local Clustering Coefficient
• Connected Components (Union Find)
• Strongly Connected Components
• Label Propagation
• Louvain Modularity
• K-1 Coloring
• Modularity Optimization
• Speaker Listener Label Propagation
Supervised
Machine Learning
• Node Classification
• Link Prediction
… and more!
Heuristic Link
Prediction
• Adamic Adar
• Common Neighbors
• Preferential Attachment
• Resource Allocations
• Same Community
• Total Neighbors
Similarity
• Node Similarity
• K-Nearest Neighbors (KNN)
• Jaccard Similarity
• Cosine Similarity
• Pearson Similarity
• Euclidean Distance
• Approximate Nearest Neighbors (ANN)
Graph
Embeddings
• Node2Vec
• FastRP
• FastRPExtended
• GraphSAGE
• Synthetic Graph Generation
• Scale Properties
• Collapse Paths
• One Hot Encoding
• Split Relationships
• Graph Export
• Pregel API (write your own algos)
https://neo4j.com/docs/graph-data-science/current/algorithms/

111
Graph Data Science Use Case Areas
Recommendations
• Personalized Recommendation
• Recommendation Engine
• Market & Customer Segmentation
• Next Best Action
Fraud & Anomaly Detection
• Risk & Safety Scoring
• First/Second/Third Party Fraud & AML
• Criminal & Other Investigative Analytics
Logistics / Supply Chain
• Performance Prediction & Analytics
• Routing & Path Finding
• Identifying Alternatives
• Risk Assessment & Dependency Mapping
Customer 360
• Entity Resolution & Master Data
Management (MDM)
• Customer Journey

112
When Do I Need Graph Algorithms?
Query (e.g. Cypher/Python)
Real-time, local decisioning
and pattern matching
Graph Algorithms
Global analysis
and iterations
You know what you’re looking
for and making a decision
You’re learning the overall structure of a
network, updating data, and predicting
Local
Patterns
Global
Computation

Going back to basics…
Unsupervised ML
• Unlabelled data
• Data driven
• Pattern identification
Supervised ML
• Labeled data
• Task driven
• Value predicting
Machine Learning
Unsupervised
Clustering
Dimension Reduction
(generalization)
Association
Data is not labeled at all Data is labeled or categorized
Divide by
similarity
Identify
sequences
Find hidden
dependencies
Stack similar clothing
Find clothes often
worn together
Make best outfit from given clothes
Supervised
Classiﬁcation
Regression
Predict a
number
Predict a
category
Predict the length
of a sock
Predict if an outfit is
fancy or casual

Going back to basics…
Unsupervised ML
• Unlabelled data
• Data driven
• Pattern identification
• Graph algorithms
Supervised ML
• Labeled data
• Task driven
• Value predicting
Community
Detection
Machine Learning
Unsupervised
Clustering
Dimension Reduction
(generalization)
Association
Data is not labeled at all Data is labeled or categorized
Divide by
similarity
Identify
sequences
Find hidden
dependencies
Which parts of my graph are
connected to each other?
Which nodes are
most similar?
How important is each node?
Supervised
Classiﬁcation
Regression
Predict a
number
Predict a
category
What’s the property value
for this node?
What type of node
is this?
Centrality
embeddings
Similarity
pathfinding

Mutable In-Memory Workspace
Computational Graph
Native Graph Store
117

GDS General Query
118

GDS Mode
119
• Stream
Returns the result of the algorithm as a stream of records.
• Stats
Returns a single record of summary statistics but does not
write to the Neo4j database or modify any data.
• Mutate
Writes the results of the algorithm to the in-memory graph
projection and returns a single record of summary statistics.
• Write
Writes the results of the algorithm back the Neo4j database
and returns a single record of summary statistics.

GDS Projection Types
120
• Native projections
Optimized for efﬁciency and performance to support graph data science at scale (Best practices).
Example:
CALL gds.graph.project('native-proj',['User', 'Movie'], ['RATED']);
• Cypher projections
Optimized for ﬂexibility and customization to support exploratory analysis, experimentation, and
smaller graph projections. Example:
CALL gds.graph.project.cypher(
'proj-cypher',
'MATCH (a:Actor) RETURN id(a) AS id, labels(a) AS labels',
'MATCH (a1:Actor)-[:ACTED_IN]->(m:Movie)<-[:ACTED_IN]-(a2)
WHERE m.year >= 1990 AND m.revenue >= 1000000
RETURN id(a1) AS source , id(a2) AS target, count(*) AS actedWithCount,
"ACTED_WITH" AS type'
);

How to use GDS
121
GDS is delivered as a library and a plugin to
Neo4j Desktop,
…or you can practice using it easily with the
Neo4j Sandbox.

Why RAG With Vector Databases Fall Short
1
3
2
4
Similarity is insufﬁcient for rich enterprise reasoning
Only leverage a fraction of
your data: Beyond simple
“metadata”, vector databases
alone fail to capture relationships
from structured data
Miss critical context: Struggle to
capture connections across
nuanced facts, making it
challenging to answer multi-step,
domain-speciﬁc, questions
Vector Similarity ≠ Relevance:
Vector search uses an incomplete
measure of similarity. Relying on it
solely can result in irrelevant and
duplicative results
Lack explainability:
The black-box nature of
vectors lacks transparency
and explainability
123 Neo4j Inc. All rights reserved 2024

GenAI
Language
Statistics
Creativity
KGs
Knowledge
Facts
Context
Accurate
Contextual
Explainable
Knowledge graphs unlock GenAI

125
RAG with Neo4j
Find similar documents
and content
Identify entities associated
to content and patterns
in connected data
Improve GenAI inferences
and insights. Discover new
relationships and entities
Unify vector search, knowledge graph and data science
capabilities to improve RAG quality and effectiveness
Vector Search Data Science
Knowledge
Graph

Improve Context
with Facts from a
Knowledge Graph
Combine native vector search with
multi-hop graph traversals to add
domain-speciﬁc context from a
knowledge graph
Incorporate hybrid retrieval, including
text search and lookups based off date,
numeric, and geopoint indexes

Use graph patterns to rank and
score vector search results
based on domain knowledge
Retrieve more reﬁned results
to power better personalized
LLM Responses
from langchain.vectorstores.neo4j_vector import
Neo4jVector
kg_personalized_search =
Neo4jVector.from_existing_index(
embedding=embedding_model,
url=NEO4J_URI,username=NEO4J_USERNAME,password=NEO4J_PA
SSWORD,
index_name='product-text-embeddings',
retrieval_query=f"""
WITH node AS product, score AS vectorSearchScore
OPTIONAL MATCH
(product) [:VARIANT_OF]-(:Article) [:PURCHASED]-
(:Customer)-[:PURCHASED] (a:Article) [:PURCHASED]-
(:Customer {{customerId: ' {CUSTOMER_ID }'}})
WITH count(a) AS purchaseScore,
RETURN
""")
Enhance
Personalization
and Quality

Expand Insights and
Inferences Using
Data Science and ML
128
Quickly incorporate context from
structured data in RAG vector search
using Neo4j graph embeddings with a
native vector index
Deeper insights for AI by enriching
your knowledge graph with link
prediction, community detection, and
classiﬁcation using graph algorithms
and machine learning
Largest Catalog of Graph
Algorithms
Native Graph
Analytics Workspace
Graph Embeddings for
Enhanced Retrieval

Interactive application to construct
knowledge graphs from PDF, youtube
transcripts etc.
Hosted on
llm-graph-builder.neo4jlabs.com
Documentation
neo4j.com/labs/genai-ecosystem/llm-graph-builder
LLM Graph Builder
Application

GraphRAG examples
GraphRAG with AWS
https://github.com/neo4j-partners/neo4j-generative-ai-aws
GraphRAG -H&M Personalized Fashion Recommendations
https://github.com/neo4j-product-examples/genai-workshop

7 Graphs in an Enterprise
131

Employees
Network & Security
Suppliers
Product
Customers
Transactions
Process
Model your data
like your business

Employees
Network & Security
Suppliers
Product
Process
Seven Graphs in
an Enterprise
Transactions
Customers

Customer
Information
History
Payments
Customers
Uncover patterns in your
Recommendations
Loyalty Programs
Churn Prevention
Customer Offers
Dynamic Pricing
Intelligent Ads
Customer Data

Customer
Information
Payments
Recommendations
Loyalty Programs
Churn Prevention
Customer Offers
Dynamic Pricing
Intelligent Ads
History
95%
improvement in
ﬁrst-call resolution
12%
increase in
Customer satisfaction
Customers
Customer Data

Payments
Product
Transactions
Customer
Invoices
Anti-money Laundering
Digital Payment Scams
Credit Risk Assessment
Claims Fraud
Credit Fraud
Transactions Data

Payments
Product
Transactions
Claims Fraud
Credit Fraud
Invoices
60%
Reduction in
infrastructure cost
69%
Improvement in
query response time
Transactions Data
Customer

Inventory
Pricing
Reviews
Product
Product Recommendations
New Product Introductions
Product Customizations
Product Inventory
Product Pricing
Product Data

Inventory
Pricing
Reviews
Product
Product Inventory
Product Pricing
Just-in-time
Price calculations for
customized products, flexible
pricing strategies without
hard-coding use cases
Product Data

Assets Location
Access
Patterns
Network & Security
Identity & Access 360
Identity Resolution
Reputation Scoring
Threat Detection
Zero Trust
Network & Security

Assets Location
Access
Patterns
4 min
to determine
blast radius
Used to take hours
or even days
100M
Customers
Protected
Network & Security
Identity Resolution
Reputation Scoring
Threat Detection
Zero Trust
Network & Security

Supplier
Inventory
Product
Suppliers
Route Planning & Optimization
Real-time Supply Chain Visibility
Inventory Planning
Risk Analysis
Suppliers Data

Supplier
Inventory
Product
10%
Reduction in
traffic
congestion
$1500
time saving per
driver
$750m/yr
Savings
Suppliers
Suppliers Data
Inventory Planning
Risk Analysis

Resources
Activity
Events
Process
Bottleneck Identiﬁcation
Process Improvement
Process Automation
Process Monitoring
Process Data

Resources
Activity
Events
Process
Process Improvement
Process Automation
Process Monitoring
83%
reduction in
query time to
quickly pinpoint
root cause of
problems
Process Data
https://neo4j.com/case-studies/boston-scientific/

Employee
Skill
s
Rol
e
Talent Learning & Development
Career Management
Orgs (who is who)
Career Development
Employees
Employee Data

Employees
Employee
Rol
e
Skill
s
Career Management
Orgs (who is who)
Career Development
5 min
From 1-2 days to enable
informed decision
making in workforce
management
Employee Data

Career Management
Orgs (who is who)
Career Development
Identity Resolution
Reputation Scoring
Threat Detection
Zero Trust
Inventory Planning
Risk Analysis
Product Inventory
Product Pricing
Process Improvement
Process Automation
Process Monitoring
Recommendations
Loyalty Programs
Churn Prevention
Customer Offers
Dynamic Pricing
Intelligent Ads
Claims Fraud
Credit Fraud
Network & Security
Suppliers
Product
Customers
Transactions
Process
Connect your graphs
to drive transformation
Employees

Cypher in Action
149

Let’s explore!
sandbox.neo4j.com
Scroll-down to:
New Project
Recommendations
Create

Question?
151

Neo4j Community Training
https://graphacademy.neo4j.com/
152

GenAI Resources
153
Exploring ChatGPT for Learning, Code, Data, NLP & Fun
Create Neo4j Database Model with ChatGPT
ChatGPT and Neo4j for Creating and Importing Sample Datasets
Doctor.ai+GPT-3+Kendra = An Ensemble Chatbot for Healthcare
Creating a Knowledge Graph From Video Transcripts With ChatGPT
Context-Aware Knowledge Graph Chatbot With GPT-4 and Neo4j
Generating Company Recommendations using LLMs & Knowledge Graphs
Integrating Neo4j into the LangChain ecosystem
Fine-tuning an LLM model with H2O LLM Studio to generate Cypher statements
Generating Cypher Queries With ChatGPT 4 on Any Graph Schema
ChatGPT Generated Star Wars Data: Let's Explore With Neo4j
LangChain has added Cypher Search
Knowledge Graphs & LLMs: Harnessing Large Language Models with Neo4j
Jump into Graph: Creating Mock Data to get started with Graph Databases
Neo4j Live: GraphGPT
OpenAI API Access - APOC Extended Documentation
LangChain Cypher Search: Tips & Tricks
Neo4j Integrations with Generative AI Features in Google Cloud Vertex AI
Neo4j Knowledge Graphs and Google Generative AI
Integrate LLM workflows with Knowledge Graph using Neo4j and APOC
Unifying Large Language Models and Knowledge Graphs: A Roadmap
Neo4j’s Role in Fueling Generative AI with Graph Technology
Knowledge Graphs & LLMs: Fine-Tuning Vs. Retrieval-Augmented Generation
Knowledge Graphs & LLMs: Multi-Hop Question Answering
Neo4j Knowledge Graphs and Google Generative AI
Personal Movie Recommendation Agent with GPT4 + Neo4j
https://github.com/neo4j-partners/neo4j-generative-ai-aws
https://github.com/neo4j-partners/neo4j-generative-ai-azure
https://github.com/neo4j-partners/neo4j-generative-ai-google-cloud
https://github.com/neo4j-partners/intelligent-app-from-videos-google-generativeai-neo4j
https://github.com/docker/genai-stack

Neo4j Professional Training
Neo4j Fundamentals
● Graphs and Graph Databases
● Basic Cypher Querying
Neo4j Developer Certification
● Data Management with Cypher
● Intermediate Cypher Querying
● Graph Data Modeling
● Data Management: Import &
Governance
● Cypher Query Tuning
Neo4j Data Scientist Certification
● Graph Data Modeling
● Data Management: Import &
Governance
● GDS Foundations
● GDS Algorithms
● GDS Monitoring & Administration
● Intro to Bloom & NeoDash
Neo4j Administrator Certification
● Basic Neo4j Administration
● Graph Dev Ops
We offer both self-paced and
instructor-led training for:
● Developers
● Data Scientists
● DBAs
Premium, in-depth content
used to train Neo4j
Professional Services
resources. Made available to
customers through a paid
subscription.
154

Thank you!
name.name@neotechnology.com
155

156
General Technical ecosystem
CI/CD
using: Jenkins, Github, ...
BI Tools:
e.g. Tableau, Tibco Spotﬁre, Qlik, …
Data Integration
Cypher Import-Bulk Import-Neo4j Data Loader
Custom API’s (Java/Python/GrapQL)-most common ETL tools (Apache HOP, KNIME, Glue, ..) -streaming data (Kafka, MQ)
-NLP: HUME, Python
Out of the box GUI:
- Neo4j Bloom
- NeoDash
- Neo4j Browser
- Linkurious
- Graphileon
- Semspect Neo4j BI Connector
Data Sources
Standard types: structured (cvs, xls, JSON, XML, RDF...) or unstructured (.doc, .pdf, …)
Data modeling
Cypher Workbench,
Arrows, Gra.fo, Yed,
Hackolade
Install/Conﬁg
Clustering, sharding, fabric
Cloud Platforms, Kubernetes, Docker
Custom GUI:
- JS: Angular, React, NodeJs, VueJs
- Java
- Structr
Drivers
Java, Python, JS, C#
Admin
using: Halin, Grafana, ...
API’s
Java, Python,GraphQL
Cypher
dev, tuning
APOC
dev, usage (Java)
GDS/NLP
dev, usage
Security
RBAC, SSO, custom Plugins
156

Neo4j CE vs Aura/EE
157

Start learning Neo4j
158
● Community edition (GPLv3),
Developer license
● Open-source software
● Built for testing and learning
● Features for Developer adoption
● Commercial license under subscription
agreement
● Closed-source software
● Built for production workloads
● Features for app owners, operational
teams, developers
Production
applications
$
Drivers and API
Transactions
and Querying
Limited analytics
capabilities
Developers
Admins
Development &
Administration
Community Edition users are graph-aware
potential enterprise customers
Buy & Deploy in Prod.

Risks of using open-source software
159
Operationally
Inefﬁcient ● Manual backups & upgrades; Downtime due to Db failure
Lack of Neo4j
prod. expertise
● Cannot implement best practices for performance
● Cannot scale analytical and transactional workloads
● No vendor oversight on source code and limited
security evaluation
Security
Copy left
obligations
● Legal implications of using open-source for proprietary
software

Who needs Commercial License?
160
Need to reduce security risk while managing conﬁdential data
Need high availability of database for critical applications
Need to achieve performance for large-scale applications
Strategic priority to move to SaaS for IT stack
● Need to safeguard competitive advantage using closed-source license
● Need professional support for project success and access to productivity tools

Who is the right customer for Aura and
Enterprise Edition?
AURA Enterprise
● AWS/ Azure/ GCP/ multi-cloud customer
● Need a SaaS solution
● Require fault tolerance and SLA for
uptime
● Fully Managed
161
Enterprise Edition
● Strategic choice to not migrate to
cloud (regulatory limitations)
● Need horizontal scaling with
federated queries on DB
secondaries
● Multiple DBs

Aura empowers enterprises with a highly secure,
reliable, and performant database for production
Operational
Reliability
Enterprise-
grade
Security
Performance
at Scale
Premium Support for project success
300+ Neo4j Engineers
162

163
Ensure Data security and integrity
User access
control
Production-
certiﬁed and
compliant
Encryption and
isolation
mechanisms
● Role based access control
● Single sign-on
● Brute force protection
● Schema-based access control
● Cypher constraints
● Data encryption (at rest, intransit)
● Intra-cluster SSL encryption
● Customer managed encryption key
● Isolated DB in VPC network with
private link
● Security evaluation of Neo4j code
(extensive testing to ensure
runtime quality)
● Vulnerability scanning
● 3rd party penetration tests
● SOC2, ISO27001, GDPR, HIPAA

Operationally reliable database for production
164
Efﬁcient DB operations with
premium support
Highly Available for Critical Apps
● 99.95% uptime with 3-server cluster
● Store back-ups across multiple
availability zones
● Backup without downtime using online
back-ups
● Point in time recovery by creating
instance from a snapshot
● Automate upgrades with no server
downtime and manual effort
● Easy to clone and pause database using
Aura console
● Tune database with visibility into metrics
through a built-in dashboard
● 24x7x365 Support from Neo4j engineers

165
Faster response times for analytical
queries (Parallel Runtime)
Trigger actions in response to
DB events (Change Data
Capture)
Monitor query performance and
error scenarios
(query logs)
Faster data retrieval with improved resource
utilization (block data format)
Improve
performance
and take
action
Improve
productivity
Visualize and explore connected data
(Bloom); collaborate with other users
Quickly upload data from multiple sources
using simple UI-tool (Data importer)
Performance at scale for queries and productivity
for users

High performance with increasing complexity and
volume of interconnected data
166
Scalable Architecture: Flexibility to vertically scale up and down Aura DB instances
I
High efficiency data ingress/ egress: export large volume of vectorized graph
data (Apache Arrow integration)
Run algorithms faster: Expand the threads for computation beyond 4 CPUs (up
to 82 CPUs)

Additional functionality in Enterprise Edition
Scale out database beyond 34 billion nodes
Partition graph across multiple databases
Retrieve data from disjointed and common graphs on multiple databases with Cypher
Scale out the read workloads by adding a new secondary server to the cluster
Multiple databases in a single account with server side routing for high availability and scalability
Scale GDS by reducing memory footprint and enabling some algorithms to run faster (BitIdMap
in-memory implementation)

Neo4j Introduction Workshop for Partners

More Related Content

Similar to Neo4j Introduction Workshop for Partners

Recently uploaded

Neo4j Introduction Workshop for Partners