SlideShare a Scribd company logo
1 of 25
Nikola Vasiljević – Head Of Insights & Analytics at Mad Head
Games
LEVERAGING SENTIMENT AND
TOPIC DETECTION FOR DECODING
PLAYER FEEDBACK
THE AIM IS TO GET YOU FAMILIAR WITH SENTIMENT AND
TOPIC ANALYSIS AND INSPIRE YOU TO USE THEM (for free) IN
YOUR DECISION-MAKING STACK.
YOU WILL LEARN:
1. INTRO TO SENTIMENT AND TOPIC
2. HOW WE UTILIZE THEM, WITH EXAMPLES
3. CHALLENGES WE FACED
4. WHAT METHODOLOGIES AND ALGORITHM WE
USE
5. KEY TAKEAWAYS WITH BENEFITS
INTRO TO SENTIMENT AND TOPIC ANALYSIS: ANALYZE YOUR PLAYERS’ REVIEW
AND COMMENTS
SENTIMENT ANALYSIS
An area of qualitative research and NLP that analyzes textual input and
assigns sentiment:
• Positive
• Neutral
• Negative
Common application: Analysis of customer satisfaction
Can be expanded to detect:
• More sentiment detail – mixed, very positive, very negative
• Specific emotions (happy, sad, thrilled, excited, angry…)
• Priority/Urgency
• Intentions (interested, not interested, purchase intent…)
QUALITATIVE DATA ANALYSIS - TOPIC DETECTION ANALYSIS
• A set of research and data analysis methods aimed
at systematically interpreting and extracting
meaningful insights from non-numerical data
• Common application: Contextual analysis of consumer
feedback
8.5
WE USE IT TO: ANALYZE PLAYERS’ COMMENTS AND REVIEWS ON THE INTERNET, AND
LEARN WHAT THEY ARE TALKING ABOUT AND HOW THEY FEEL ABOUT IT
This enables us to deliver insights on how we can make and improve games based on
players’ feedback
• Use main genre drivers as pillars
• Make different game design decision
• Solve bugs
• Balance out our games
• Refine our communication to manage players’ expectations
• Patch up our games
USER ATTITUDE
AND SATISFACTION
ANALYZE
COMPETITION
PLAYER
SEGMENTATION
ANALYZE GENRES’
MAIN DRIVERS
CHALLENGES: PROVIDE INSIGHTFUL AND TIMELY FEEDBACK DURING THE STEPS OF THE
PRODUCTION CYCLE
TRADITIONAL QUALITATIVE METHODS
• Require unbiased experts
• Take a lot of time
• Hard to scale for big data
• Cannot be fully automated
ML – NLP and LLMs
• Understanding irony, double meaning,
sarcasm
• Broken language and mistakes
• Industry-specific slang – n00b, OP, take my
money…
• Understanding context (“Andromeda anyone?“)
To get useful, actionable
insights, it requires:
• a mixture of methods
• a lot of fine tuning
• a human “bus driver“
• CHALLENGING, IN A POSITIVE WAY, AND FUN
• ELEMENTAL WEAPONS AND ENVIRONMENTAL
DAMAGE SYSTEM IS COOL
• BOSS FIGHTS ARE EXCITING
• LEARNING ATTACK PATTERNS IS REWARDING
• REQUIRES SKILL, WHICH IS REWARDING
• UNFAIR AND IMBALANCED COMBAT
• BOSS DIFFICULTY, ESPECIALLY THE FIRST ONE IS
UNFAIR
• NO PILLAR TO SAVE THE GAME BEFORE THE
BOSS FIGHT
• WAS THERE FOR THE STORY, BUT THE COMBAT
IS TOO MUCH
• BOSS DIFFICULTY SPIKE
• WHY IS IT SOULSIKE
WHAT
NEXT?
EXAMPLE FROM SCARS ABOVE: A TOP DOWN ANALYSIS OF PLAYERS’
SENTIMENT AND TOPIC DETECTION
Sub-topic
Sub-topic
GROUNDED THEORY: FORM THEORIES AND HYPOTHESES BASED ON DATA, TEST
THEM AND TAKE ACTION
What we did
• Conducted additional research to find out the nature and the differences between
story-driven players and combat-driven players, which gave us confirmation for our
hypothesis
• Lowered overall difficulty
• Adjusted the design of some boss fights
• Reworked the difficulty settings by making game design changes to the EASY mode 
removed some souslike elements, like respawning enemies
ADJUSTING COMBAT / BOSS DIFFICULTY BY MAKING IT EASIER WILL MAKE THE GAME
PLAYABLE FOR NARRATIVE-FIRST PLAYERS WHILE NOT HARMING THE EXPERIENCE OF THE
SOULSLIKE PLAYERS TARGET SUB-GROUP
Why is it important to develop theories?
• Game design elements are intertwined and work in balance – tweaking one will
inevitably impact others, so taking on them separately will get you nowhere
• We have multiple target sub-groups which sometimes are driven by totally opposite
drivers  too much catering to one may repel another
SENTIMENT ANALYSIS: A MIX OF APPROACHES
SENTIMENT ANALYSIS: A MIX OF APPROACHES
1. Devise a sample
(N=1000)
2. Go manually through
each review/comment
and assign sentiment
3. Use it as a training
model
SENTIMENT ANALYSIS: A MIX OF APPROACHES
1. NLTK Python library
2. Calculate sentiment for each
review/comment
3. We used a pre-trained model
with fine tuning
Use OpenAI API to make your
own assistant which will do
sentiment analysis for you
SENTIMENT ANALYSIS: A MIX OF APPROACHES
NLTK OVERVIEW: A PRE-TRAINED MODEL WITH FINE TUNING WORKS SOLIDLY
WELL
1. Text preprocessing
• Remove stop-words (words of little value
(the, and, is, in, of…)
• Word exclusion – if there are any
particular words to exclude
• Tokenization – the text is transformed
into individual tokens to facilitate
analysis at a more granular level
• Stemming – removing suffixes to get the
root of the word (run, running, runs run)
• Lemmatization - reducing words to their
base or dictionary form, known as the
"lemma." It considers the meaning of the
word and applies morphological analysis to
obtain the root form; good, better, best 
good
2. Sentiment intensity analyzer
• Apply sentiment scores – calculates
sentiment for textual input based on
tokens’ sentiment scores
3. Fine-tuning
• Manually adjusting sentiment thresholds
• Adding a custom dictionary (“sick, OP,
take my money…“)
Comment Sentiment
0 Holy hell. That was a movie-level trailer!<br>... Positive
1 Animations are so bad seeing the graphics it d... Negative
2 Pyramids and darkness? Destiny is leaking Neutral
3 <b>facepalm</b> the running animation is still... Neutral
4 This looks interesting Positive
.. ... ...
111 oh again STONG women? facepalm Negative
112 Alles sehr schön. Aber zuerst zusammen die Num... Positive
113 Let&#39;s goooo Positive
114 Awesome: I&#39;m eager for this game Positive
115 Really looking forward to this game! Positive
THE OUTPUT
ASSIGNING SENTIMENT
SCORES
OPEN AI: PROMPT-ENGINEER YOUR OWN ASSISTANT
1. Specify model (gpt 3.5 turbo, gpt-4…)
2. Define messages:
1. Role of the messenger: system, user, assistant
2. Content of the message – tell it what it should do
Prompt engineering step-by-step*:
a) system message
b) very specific instructions, each given in a separate prompt
c) Provide input data
d) Add contextual information
e) Set temperature (controls randomness of the output)
f) Specify desired output format (json, csv)
* Goran Milovanović: Generative AI in Action: Three case studies with OpenAI GPT(3-3.5)
THE
OUTPUT
DEFINE YOUR ASSISTANT
TOPIC ANALYSIS: A MIX OF APPROACHES
TOPIC ANALYSIS: A MIX OF APPROACHES
1. Devise a sample
(N=1000)
2. Go manually through
each review/comment
and assign tags
3. Synthesis – group
tags into subtopics
and topics
TOPIC ANALYSIS: A MIX OF APPROACHES
1. Take the keywords
devised with qualitative
coding
2. Setup IF THEN ELSE
classification rules for
keyword search of your
text input, and group by
topic and subtopic
TOPIC ANALYSIS: A MIX OF APPROACHES
BERT AND LDA on their
own didn’t get us to where
we wanted.
LDA + OPEN AI LLM:
1. Preprocess the data (as
in NLTK)
2. Extract topics and
associeated words with
LDA to a list
3. Take the extracted
topics and generate a
more user-friendly
output
TOPIC ANALYSIS: A MIX OF APPROACHES
Use OpenAI API to make
your own assistant which
will do sentiment analysis
for you
QUALITATIVE CODING AND ASPECT BASED TOPIC DETECTION: MANUAL WORK
WITH HIGH REWARD
COMBAT DIFFICULTY
HITTING WEAKSPOTS
FIRE WEAPON RELOAD
SHORT WEAKSPOT
OPPORTUNITY WINDOW
..It was sooo unfair hitting the glowing part of the chest with the fire weapon, as reload times are
too long so you miss the opporutnity window…
COMBAT DIFFICULTY
WEAPONS
COMBAT DESIGN
COMBAT DIFFICULTY
SHOOTING
RELOAD
COMBAT DESIGN
1. Take the keywords devised with qualitative
coding
2. Setup IF THEN ELSE or REGEX classification
rules for keyword search of text input,
and group by topic&subtopic
ASPECT BASED HARD-CODED TOPIC DETECTION
LDA + OPEN AI LLM: A WELL-ROUNDED IMPRESSIONS SUMMARY…
1. Define a function (A) to extract topics and associated
keywords from the input
2. Provide it parameters
1. Input parameters
2. Number of topics to discover
3. Number of words to include per topic
4. Output parameters – a list of topics with sublists
of keywords
3. Define a function (B) to take the previous output and
generate descriptive texts about detected topics using
detected keywords
4. Define parameters
1. LLM
2. Input parameters
3. Number of topics
4. Number of words per topic
5. Output format
OUTPUT
FUNCTION A
FUNCTION B
(a)
(b)
OPEN AI TOPIC DETECTION: UNDERSTANDS CONTEXT BUT LIKES TO TALK IN
SYNONYMS AND IS OCCASIONALLY STUBBORN
1. Specify model (gpt 3.5 turbo, gpt-4…)
2. Define messages:
1. Role of the messenger: system, user, assistant
2. Content of the message – tell it what it should do
Prompt engineering step-by-step*:
a) system message
b) very specific instructions, each given in a separate prompt
c) Provide input data
d) Add contextual information
e) Set temperature (controls randomness of the output)
f) Specify desired output format (json, csv)
DEFINE YOUR ASSISTANT
OUTPUT
Comment Sentiment Topic 1 Topic 2 Topic 3 Topic 4 Topic 5
0 Holy hell. That was a movie-level trailer!<br>... Positive Appreciation Trailer Comparison Quality
1 Animations are so bad seeing the graphics it d... Negative Animation Graphics Quality
2 Pyramids and darkness? Destiny is leaking Neutral Comparison Darkness
3 <b>facepalm</b> the running animation is still... Mixed Animation Appreciation
4 This looks interesting Positive Interest Appearance
.. ... ...
111 oh again STONG women? facepalm Negative Character Face
112 Alles sehr schön. Aber zuerst zusammen die Num... Mixed Potential
113 Let&#39;s goooo Positive Movement
114 Awesome: I&#39;m eager for this game Positive Interest
115 Really looking forward to this game! Positive Anticipation
[116 rows x 2 columns]
A HOLISTIC APPROACH TO ANALYZING PLAYERS’S FEEDBACK SENTIMENT AND
TOPIC: HOW WE DO IT
INITIAL MANUAL SENTIMENT AND TOPIC ANALYSIS USING THE QUALITATIVE CODING METHODOLOGY
• Devise a N=1000 sample, and perform a manual sentiment and topic analysis with qualitative coding
• Use it as a training dataset for ML
• Listed topics, subtopics and keywords are used as input instructions for aspect-based topic
analysis, as well as for ML fine tuning
• You get to really acquaint yourself with the data and get to know feeling, the players, the persons
behind them  you get into players’ shoes
ASPECT BASED TOPIC ANALYSIS:
• ML/AI Sentiment analysis (fine tuned using the aspects from previous step)
• Use the aspects and keywords to form keywordssubtopicstopics and fine tune your
tool/algorithm to look for them (you don’t need ML for this, a simple IF THEN ELSE to count
topics and subtopics mentioned based on keywords will do (we do this in Tableau)
• It’s all-encompassing
LLM AND OPEN AI MODELS
• Compare results
• Validate your findings from aspect-based models
• Enhance the previous results – something may pop up that you didn’t encompass with your
manually defined aspects
• Keep fine tuning
GROUNDED THEORY  CONNECTING THE DOTS BY DEVELOPING INSIGHTS AND HYPOTHESES BASED ON DATA
• It aims to develop theories or concepts grounded in the data, allowing patterns and themes to
emerge organically from the data without preconceived notions.
YOU CAN BUILD YOUR OWN AUTOMATED TOOLS: THIS IS OUR COMPETITION
ANALYSIS TOOL
YouTube Scrapping
script
NLTK (fine-
tuned)
Aspect-based
KEY TAKEAWAYS
USE SENTIMENT AND TOPIC FOR
PLAYER FEEDBACK ANALYSIS
• There’s an abundance of free data
• You will learn a lot about your players
• It WILL help you make your games better
USE THE DATA TO GET TO KNOW
YOUR PLAYERS
• Understand your players and develop
segments
• Understand that you have many target
subgroups
• Understand what are the main drivers for
each of them
• But, also understand you can’t please
everyone
DEVELOP YOUR OWN
FRAMEWORK
• Experiment with all methodologies, don’t discriminate
• Traditional (qualitative coding)
• Automated (ML algorithms, OpenAI)
• Combine them and find out what works for you
DEVELOP INSIGHTS USING
GROUNDED THEORY
MAKE IT A PART OF YOUR DECISION-
MAKING STACK
DO THE MANUAL WORK
• Take the time and do an initial manual analysis on N=1000
reviews or comments
• It will familiarize you with the data
• Use it as a training set
• Go beyond plain data and develop theories with your game
designer, based on the data and knowledge you acquired – not
just with Sentiment and Topic analysis, but with all of your
insights
• Test the hypotheses
Analyze your game life-cycle and find out how and when it can be
used in to make decisions and connect with your decision makers:
• During production
• During soft-launch / demo
• When a game is live
THANK
My contact info:
vasiljevic.d.nikola@gmail.com
nikola.vasiljevic@madheadgames.com
Add me on LinkedIn via QR code:

More Related Content

Similar to Decode Player Feedback Using Sentiment and Topic Analysis

Vulnerability, exploit to metasploit
Vulnerability, exploit to metasploitVulnerability, exploit to metasploit
Vulnerability, exploit to metasploitTiago Henriques
 
Five Cliches of Online Game Development
Five Cliches of Online Game DevelopmentFive Cliches of Online Game Development
Five Cliches of Online Game Developmentiandundore
 
Adaptive Games Content Generation - 2D Mario
Adaptive Games Content Generation - 2D MarioAdaptive Games Content Generation - 2D Mario
Adaptive Games Content Generation - 2D MarioMohammad Shaker
 
Final_Presentation
Final_PresentationFinal_Presentation
Final_PresentationJoe Totherow
 
Mirko Lucchese - Deep Image Processing
Mirko Lucchese - Deep Image ProcessingMirko Lucchese - Deep Image Processing
Mirko Lucchese - Deep Image ProcessingMeetupDataScienceRoma
 
How to Design Effective Learning Games: Sharon Boller and Karl Kapp
How to Design Effective Learning Games: Sharon Boller and Karl KappHow to Design Effective Learning Games: Sharon Boller and Karl Kapp
How to Design Effective Learning Games: Sharon Boller and Karl KappSharon Boller
 
Understanding and improving games through machine learning - Natasha Latysheva
Understanding and improving games through machine learning - Natasha LatyshevaUnderstanding and improving games through machine learning - Natasha Latysheva
Understanding and improving games through machine learning - Natasha LatyshevaLauren Cormack
 
How Sentiment Analysis works
How Sentiment Analysis worksHow Sentiment Analysis works
How Sentiment Analysis worksCJ Jenkins
 
Sentiment Analysis
Sentiment AnalysisSentiment Analysis
Sentiment AnalysisSagar Ahire
 
Structure And Types Of Scientific Writing Free Ess
Structure And Types Of Scientific Writing Free EssStructure And Types Of Scientific Writing Free Ess
Structure And Types Of Scientific Writing Free EssWendy Fox
 
Answering question 1
Answering question 1Answering question 1
Answering question 1sssfcmedia
 
LAFS SVI Level 3 - Game Design and Analysis
LAFS SVI Level 3 - Game Design and AnalysisLAFS SVI Level 3 - Game Design and Analysis
LAFS SVI Level 3 - Game Design and AnalysisDavid Mullich
 
Pycon India 2018 Natural Language Processing Workshop
Pycon India 2018   Natural Language Processing WorkshopPycon India 2018   Natural Language Processing Workshop
Pycon India 2018 Natural Language Processing WorkshopLakshya Sivaramakrishnan
 

Similar to Decode Player Feedback Using Sentiment and Topic Analysis (20)

Vulnerability, exploit to metasploit
Vulnerability, exploit to metasploitVulnerability, exploit to metasploit
Vulnerability, exploit to metasploit
 
Five Cliches of Online Game Development
Five Cliches of Online Game DevelopmentFive Cliches of Online Game Development
Five Cliches of Online Game Development
 
Adaptive Games Content Generation - 2D Mario
Adaptive Games Content Generation - 2D MarioAdaptive Games Content Generation - 2D Mario
Adaptive Games Content Generation - 2D Mario
 
Final_Presentation
Final_PresentationFinal_Presentation
Final_Presentation
 
Mirko Lucchese - Deep Image Processing
Mirko Lucchese - Deep Image ProcessingMirko Lucchese - Deep Image Processing
Mirko Lucchese - Deep Image Processing
 
How to Design Effective Learning Games: Sharon Boller and Karl Kapp
How to Design Effective Learning Games: Sharon Boller and Karl KappHow to Design Effective Learning Games: Sharon Boller and Karl Kapp
How to Design Effective Learning Games: Sharon Boller and Karl Kapp
 
InternshipReport
InternshipReportInternshipReport
InternshipReport
 
01 - Introduction to Game Mechanics
01 - Introduction to Game Mechanics01 - Introduction to Game Mechanics
01 - Introduction to Game Mechanics
 
Understanding and improving games through machine learning - Natasha Latysheva
Understanding and improving games through machine learning - Natasha LatyshevaUnderstanding and improving games through machine learning - Natasha Latysheva
Understanding and improving games through machine learning - Natasha Latysheva
 
How Sentiment Analysis works
How Sentiment Analysis worksHow Sentiment Analysis works
How Sentiment Analysis works
 
Sentiment Analysis
Sentiment AnalysisSentiment Analysis
Sentiment Analysis
 
Creative Engineering 101
Creative Engineering 101Creative Engineering 101
Creative Engineering 101
 
Structure And Types Of Scientific Writing Free Ess
Structure And Types Of Scientific Writing Free EssStructure And Types Of Scientific Writing Free Ess
Structure And Types Of Scientific Writing Free Ess
 
U1 Lesson 06
U1 Lesson 06U1 Lesson 06
U1 Lesson 06
 
Gamification
GamificationGamification
Gamification
 
Answering question 1
Answering question 1Answering question 1
Answering question 1
 
LAFS SVI Level 3 - Game Design and Analysis
LAFS SVI Level 3 - Game Design and AnalysisLAFS SVI Level 3 - Game Design and Analysis
LAFS SVI Level 3 - Game Design and Analysis
 
Core gameplay analysis
Core gameplay analysisCore gameplay analysis
Core gameplay analysis
 
Pycon India 2018 Natural Language Processing Workshop
Pycon India 2018   Natural Language Processing WorkshopPycon India 2018   Natural Language Processing Workshop
Pycon India 2018 Natural Language Processing Workshop
 
NLP Bootcamp
NLP BootcampNLP Bootcamp
NLP Bootcamp
 

More from DataScienceConferenc1

[DSC MENA 24] Mostafa_Essa_-_Ai_and_cloud.pdf
[DSC MENA 24] Mostafa_Essa_-_Ai_and_cloud.pdf[DSC MENA 24] Mostafa_Essa_-_Ai_and_cloud.pdf
[DSC MENA 24] Mostafa_Essa_-_Ai_and_cloud.pdfDataScienceConferenc1
 
[DSC MENA 24] Yasser_El_Bendary - How NLP & LLMs model can excel in comprehen...
[DSC MENA 24] Yasser_El_Bendary - How NLP & LLMs model can excel in comprehen...[DSC MENA 24] Yasser_El_Bendary - How NLP & LLMs model can excel in comprehen...
[DSC MENA 24] Yasser_El_Bendary - How NLP & LLMs model can excel in comprehen...DataScienceConferenc1
 
[DSC MENA 24] Medhat_Kandil - Empowering Egypt's AI & Biotechnology Scenes.pdf
[DSC MENA 24] Medhat_Kandil - Empowering Egypt's AI & Biotechnology Scenes.pdf[DSC MENA 24] Medhat_Kandil - Empowering Egypt's AI & Biotechnology Scenes.pdf
[DSC MENA 24] Medhat_Kandil - Empowering Egypt's AI & Biotechnology Scenes.pdfDataScienceConferenc1
 
[DSC MENA 24] Youssef_Kamal - Data governance and quality.pdf
[DSC MENA 24] Youssef_Kamal - Data governance and quality.pdf[DSC MENA 24] Youssef_Kamal - Data governance and quality.pdf
[DSC MENA 24] Youssef_Kamal - Data governance and quality.pdfDataScienceConferenc1
 
[DSC MENA 24] Abdelrahman_Ghallab_-_Data_Product_mgmt.pdf
[DSC MENA 24] Abdelrahman_Ghallab_-_Data_Product_mgmt.pdf[DSC MENA 24] Abdelrahman_Ghallab_-_Data_Product_mgmt.pdf
[DSC MENA 24] Abdelrahman_Ghallab_-_Data_Product_mgmt.pdfDataScienceConferenc1
 
[DSC MENA 24] Asmaa_Eltaher_-_Innovation_Beyond_Brainstorming.pptx
[DSC MENA 24] Asmaa_Eltaher_-_Innovation_Beyond_Brainstorming.pptx[DSC MENA 24] Asmaa_Eltaher_-_Innovation_Beyond_Brainstorming.pptx
[DSC MENA 24] Asmaa_Eltaher_-_Innovation_Beyond_Brainstorming.pptxDataScienceConferenc1
 
[DSC MENA 24] Muhammad_Ezzat_-_Sustianable_Growth_Empowerment.pdf
[DSC MENA 24] Muhammad_Ezzat_-_Sustianable_Growth_Empowerment.pdf[DSC MENA 24] Muhammad_Ezzat_-_Sustianable_Growth_Empowerment.pdf
[DSC MENA 24] Muhammad_Ezzat_-_Sustianable_Growth_Empowerment.pdfDataScienceConferenc1
 
[DSC MENA 24] Basma_Rady_-_Building_a_Data_Driven_Culture_in_Your_Organizatio...
[DSC MENA 24] Basma_Rady_-_Building_a_Data_Driven_Culture_in_Your_Organizatio...[DSC MENA 24] Basma_Rady_-_Building_a_Data_Driven_Culture_in_Your_Organizatio...
[DSC MENA 24] Basma_Rady_-_Building_a_Data_Driven_Culture_in_Your_Organizatio...DataScienceConferenc1
 
[DSC MENA 24] Ahmed_Muselhy_-_Unveiling-the-Secrets-of-AI-in-Hiring.pdf
[DSC MENA 24] Ahmed_Muselhy_-_Unveiling-the-Secrets-of-AI-in-Hiring.pdf[DSC MENA 24] Ahmed_Muselhy_-_Unveiling-the-Secrets-of-AI-in-Hiring.pdf
[DSC MENA 24] Ahmed_Muselhy_-_Unveiling-the-Secrets-of-AI-in-Hiring.pdfDataScienceConferenc1
 
[DSC MENA 24] Ziad_Diab_-_Data-Driven_Disruption_-_The_Role_of_Data_Strategy_...
[DSC MENA 24] Ziad_Diab_-_Data-Driven_Disruption_-_The_Role_of_Data_Strategy_...[DSC MENA 24] Ziad_Diab_-_Data-Driven_Disruption_-_The_Role_of_Data_Strategy_...
[DSC MENA 24] Ziad_Diab_-_Data-Driven_Disruption_-_The_Role_of_Data_Strategy_...DataScienceConferenc1
 
[DSC MENA 24] Mohammad_Essam_- Leveraging Scene Graphs for Generative AI and ...
[DSC MENA 24] Mohammad_Essam_- Leveraging Scene Graphs for Generative AI and ...[DSC MENA 24] Mohammad_Essam_- Leveraging Scene Graphs for Generative AI and ...
[DSC MENA 24] Mohammad_Essam_- Leveraging Scene Graphs for Generative AI and ...DataScienceConferenc1
 
[DSC MENA 24] Ahmed_Fahmy - Navigating the Future.pdf
[DSC MENA 24] Ahmed_Fahmy - Navigating the Future.pdf[DSC MENA 24] Ahmed_Fahmy - Navigating the Future.pdf
[DSC MENA 24] Ahmed_Fahmy - Navigating the Future.pdfDataScienceConferenc1
 
[DSC MENA 24] Hany_Saad_Gheit_-_Azure_OpenAI_service.pptx
[DSC MENA 24] Hany_Saad_Gheit_-_Azure_OpenAI_service.pptx[DSC MENA 24] Hany_Saad_Gheit_-_Azure_OpenAI_service.pptx
[DSC MENA 24] Hany_Saad_Gheit_-_Azure_OpenAI_service.pptxDataScienceConferenc1
 
[DSC MENA 24] Nezar_El_Kady_-_From_Turing_to_Transformers__Navigating_the_AI_...
[DSC MENA 24] Nezar_El_Kady_-_From_Turing_to_Transformers__Navigating_the_AI_...[DSC MENA 24] Nezar_El_Kady_-_From_Turing_to_Transformers__Navigating_the_AI_...
[DSC MENA 24] Nezar_El_Kady_-_From_Turing_to_Transformers__Navigating_the_AI_...DataScienceConferenc1
 
[DSC MENA 24] Amira_Abdelaziz_-_AI_in_Financial_Services.pptx
[DSC MENA 24] Amira_Abdelaziz_-_AI_in_Financial_Services.pptx[DSC MENA 24] Amira_Abdelaziz_-_AI_in_Financial_Services.pptx
[DSC MENA 24] Amira_Abdelaziz_-_AI_in_Financial_Services.pptxDataScienceConferenc1
 
[DSC MENA 24] Omar_Ossama - My Journey from the Field of Oil & Gas, to the Ex...
[DSC MENA 24] Omar_Ossama - My Journey from the Field of Oil & Gas, to the Ex...[DSC MENA 24] Omar_Ossama - My Journey from the Field of Oil & Gas, to the Ex...
[DSC MENA 24] Omar_Ossama - My Journey from the Field of Oil & Gas, to the Ex...DataScienceConferenc1
 
[DSC MENA 24] Ramy_Agieb_-_Advancements_in_Artificial_Intelligence_for_Cybers...
[DSC MENA 24] Ramy_Agieb_-_Advancements_in_Artificial_Intelligence_for_Cybers...[DSC MENA 24] Ramy_Agieb_-_Advancements_in_Artificial_Intelligence_for_Cybers...
[DSC MENA 24] Ramy_Agieb_-_Advancements_in_Artificial_Intelligence_for_Cybers...DataScienceConferenc1
 
[DSC MENA 24] Sohaila_Diab_-_Lets_Talk_Gen_AI_Presentation.pptx
[DSC MENA 24] Sohaila_Diab_-_Lets_Talk_Gen_AI_Presentation.pptx[DSC MENA 24] Sohaila_Diab_-_Lets_Talk_Gen_AI_Presentation.pptx
[DSC MENA 24] Sohaila_Diab_-_Lets_Talk_Gen_AI_Presentation.pptxDataScienceConferenc1
 
[DSC MENA 24] Amal_Elgammal_-_QUALITOP_presentation.pptx
[DSC MENA 24] Amal_Elgammal_-_QUALITOP_presentation.pptx[DSC MENA 24] Amal_Elgammal_-_QUALITOP_presentation.pptx
[DSC MENA 24] Amal_Elgammal_-_QUALITOP_presentation.pptxDataScienceConferenc1
 
[DSC MENA 24] Abdelrahman_Sleem_-_AI_For_Marketing_DSC.pdf
[DSC MENA 24] Abdelrahman_Sleem_-_AI_For_Marketing_DSC.pdf[DSC MENA 24] Abdelrahman_Sleem_-_AI_For_Marketing_DSC.pdf
[DSC MENA 24] Abdelrahman_Sleem_-_AI_For_Marketing_DSC.pdfDataScienceConferenc1
 

More from DataScienceConferenc1 (20)

[DSC MENA 24] Mostafa_Essa_-_Ai_and_cloud.pdf
[DSC MENA 24] Mostafa_Essa_-_Ai_and_cloud.pdf[DSC MENA 24] Mostafa_Essa_-_Ai_and_cloud.pdf
[DSC MENA 24] Mostafa_Essa_-_Ai_and_cloud.pdf
 
[DSC MENA 24] Yasser_El_Bendary - How NLP & LLMs model can excel in comprehen...
[DSC MENA 24] Yasser_El_Bendary - How NLP & LLMs model can excel in comprehen...[DSC MENA 24] Yasser_El_Bendary - How NLP & LLMs model can excel in comprehen...
[DSC MENA 24] Yasser_El_Bendary - How NLP & LLMs model can excel in comprehen...
 
[DSC MENA 24] Medhat_Kandil - Empowering Egypt's AI & Biotechnology Scenes.pdf
[DSC MENA 24] Medhat_Kandil - Empowering Egypt's AI & Biotechnology Scenes.pdf[DSC MENA 24] Medhat_Kandil - Empowering Egypt's AI & Biotechnology Scenes.pdf
[DSC MENA 24] Medhat_Kandil - Empowering Egypt's AI & Biotechnology Scenes.pdf
 
[DSC MENA 24] Youssef_Kamal - Data governance and quality.pdf
[DSC MENA 24] Youssef_Kamal - Data governance and quality.pdf[DSC MENA 24] Youssef_Kamal - Data governance and quality.pdf
[DSC MENA 24] Youssef_Kamal - Data governance and quality.pdf
 
[DSC MENA 24] Abdelrahman_Ghallab_-_Data_Product_mgmt.pdf
[DSC MENA 24] Abdelrahman_Ghallab_-_Data_Product_mgmt.pdf[DSC MENA 24] Abdelrahman_Ghallab_-_Data_Product_mgmt.pdf
[DSC MENA 24] Abdelrahman_Ghallab_-_Data_Product_mgmt.pdf
 
[DSC MENA 24] Asmaa_Eltaher_-_Innovation_Beyond_Brainstorming.pptx
[DSC MENA 24] Asmaa_Eltaher_-_Innovation_Beyond_Brainstorming.pptx[DSC MENA 24] Asmaa_Eltaher_-_Innovation_Beyond_Brainstorming.pptx
[DSC MENA 24] Asmaa_Eltaher_-_Innovation_Beyond_Brainstorming.pptx
 
[DSC MENA 24] Muhammad_Ezzat_-_Sustianable_Growth_Empowerment.pdf
[DSC MENA 24] Muhammad_Ezzat_-_Sustianable_Growth_Empowerment.pdf[DSC MENA 24] Muhammad_Ezzat_-_Sustianable_Growth_Empowerment.pdf
[DSC MENA 24] Muhammad_Ezzat_-_Sustianable_Growth_Empowerment.pdf
 
[DSC MENA 24] Basma_Rady_-_Building_a_Data_Driven_Culture_in_Your_Organizatio...
[DSC MENA 24] Basma_Rady_-_Building_a_Data_Driven_Culture_in_Your_Organizatio...[DSC MENA 24] Basma_Rady_-_Building_a_Data_Driven_Culture_in_Your_Organizatio...
[DSC MENA 24] Basma_Rady_-_Building_a_Data_Driven_Culture_in_Your_Organizatio...
 
[DSC MENA 24] Ahmed_Muselhy_-_Unveiling-the-Secrets-of-AI-in-Hiring.pdf
[DSC MENA 24] Ahmed_Muselhy_-_Unveiling-the-Secrets-of-AI-in-Hiring.pdf[DSC MENA 24] Ahmed_Muselhy_-_Unveiling-the-Secrets-of-AI-in-Hiring.pdf
[DSC MENA 24] Ahmed_Muselhy_-_Unveiling-the-Secrets-of-AI-in-Hiring.pdf
 
[DSC MENA 24] Ziad_Diab_-_Data-Driven_Disruption_-_The_Role_of_Data_Strategy_...
[DSC MENA 24] Ziad_Diab_-_Data-Driven_Disruption_-_The_Role_of_Data_Strategy_...[DSC MENA 24] Ziad_Diab_-_Data-Driven_Disruption_-_The_Role_of_Data_Strategy_...
[DSC MENA 24] Ziad_Diab_-_Data-Driven_Disruption_-_The_Role_of_Data_Strategy_...
 
[DSC MENA 24] Mohammad_Essam_- Leveraging Scene Graphs for Generative AI and ...
[DSC MENA 24] Mohammad_Essam_- Leveraging Scene Graphs for Generative AI and ...[DSC MENA 24] Mohammad_Essam_- Leveraging Scene Graphs for Generative AI and ...
[DSC MENA 24] Mohammad_Essam_- Leveraging Scene Graphs for Generative AI and ...
 
[DSC MENA 24] Ahmed_Fahmy - Navigating the Future.pdf
[DSC MENA 24] Ahmed_Fahmy - Navigating the Future.pdf[DSC MENA 24] Ahmed_Fahmy - Navigating the Future.pdf
[DSC MENA 24] Ahmed_Fahmy - Navigating the Future.pdf
 
[DSC MENA 24] Hany_Saad_Gheit_-_Azure_OpenAI_service.pptx
[DSC MENA 24] Hany_Saad_Gheit_-_Azure_OpenAI_service.pptx[DSC MENA 24] Hany_Saad_Gheit_-_Azure_OpenAI_service.pptx
[DSC MENA 24] Hany_Saad_Gheit_-_Azure_OpenAI_service.pptx
 
[DSC MENA 24] Nezar_El_Kady_-_From_Turing_to_Transformers__Navigating_the_AI_...
[DSC MENA 24] Nezar_El_Kady_-_From_Turing_to_Transformers__Navigating_the_AI_...[DSC MENA 24] Nezar_El_Kady_-_From_Turing_to_Transformers__Navigating_the_AI_...
[DSC MENA 24] Nezar_El_Kady_-_From_Turing_to_Transformers__Navigating_the_AI_...
 
[DSC MENA 24] Amira_Abdelaziz_-_AI_in_Financial_Services.pptx
[DSC MENA 24] Amira_Abdelaziz_-_AI_in_Financial_Services.pptx[DSC MENA 24] Amira_Abdelaziz_-_AI_in_Financial_Services.pptx
[DSC MENA 24] Amira_Abdelaziz_-_AI_in_Financial_Services.pptx
 
[DSC MENA 24] Omar_Ossama - My Journey from the Field of Oil & Gas, to the Ex...
[DSC MENA 24] Omar_Ossama - My Journey from the Field of Oil & Gas, to the Ex...[DSC MENA 24] Omar_Ossama - My Journey from the Field of Oil & Gas, to the Ex...
[DSC MENA 24] Omar_Ossama - My Journey from the Field of Oil & Gas, to the Ex...
 
[DSC MENA 24] Ramy_Agieb_-_Advancements_in_Artificial_Intelligence_for_Cybers...
[DSC MENA 24] Ramy_Agieb_-_Advancements_in_Artificial_Intelligence_for_Cybers...[DSC MENA 24] Ramy_Agieb_-_Advancements_in_Artificial_Intelligence_for_Cybers...
[DSC MENA 24] Ramy_Agieb_-_Advancements_in_Artificial_Intelligence_for_Cybers...
 
[DSC MENA 24] Sohaila_Diab_-_Lets_Talk_Gen_AI_Presentation.pptx
[DSC MENA 24] Sohaila_Diab_-_Lets_Talk_Gen_AI_Presentation.pptx[DSC MENA 24] Sohaila_Diab_-_Lets_Talk_Gen_AI_Presentation.pptx
[DSC MENA 24] Sohaila_Diab_-_Lets_Talk_Gen_AI_Presentation.pptx
 
[DSC MENA 24] Amal_Elgammal_-_QUALITOP_presentation.pptx
[DSC MENA 24] Amal_Elgammal_-_QUALITOP_presentation.pptx[DSC MENA 24] Amal_Elgammal_-_QUALITOP_presentation.pptx
[DSC MENA 24] Amal_Elgammal_-_QUALITOP_presentation.pptx
 
[DSC MENA 24] Abdelrahman_Sleem_-_AI_For_Marketing_DSC.pdf
[DSC MENA 24] Abdelrahman_Sleem_-_AI_For_Marketing_DSC.pdf[DSC MENA 24] Abdelrahman_Sleem_-_AI_For_Marketing_DSC.pdf
[DSC MENA 24] Abdelrahman_Sleem_-_AI_For_Marketing_DSC.pdf
 

Recently uploaded

Halmar dropshipping via API with DroFx
Halmar  dropshipping  via API with DroFxHalmar  dropshipping  via API with DroFx
Halmar dropshipping via API with DroFxolyaivanovalion
 
Introduction-to-Machine-Learning (1).pptx
Introduction-to-Machine-Learning (1).pptxIntroduction-to-Machine-Learning (1).pptx
Introduction-to-Machine-Learning (1).pptxfirstjob4
 
Vip Model Call Girls (Delhi) Karol Bagh 9711199171✔️Body to body massage wit...
Vip Model  Call Girls (Delhi) Karol Bagh 9711199171✔️Body to body massage wit...Vip Model  Call Girls (Delhi) Karol Bagh 9711199171✔️Body to body massage wit...
Vip Model Call Girls (Delhi) Karol Bagh 9711199171✔️Body to body massage wit...shivangimorya083
 
꧁❤ Greater Noida Call Girls Delhi ❤꧂ 9711199171 ☎️ Hard And Sexy Vip Call
꧁❤ Greater Noida Call Girls Delhi ❤꧂ 9711199171 ☎️ Hard And Sexy Vip Call꧁❤ Greater Noida Call Girls Delhi ❤꧂ 9711199171 ☎️ Hard And Sexy Vip Call
꧁❤ Greater Noida Call Girls Delhi ❤꧂ 9711199171 ☎️ Hard And Sexy Vip Callshivangimorya083
 
BPAC WITH UFSBI GENERAL PRESENTATION 18_05_2017-1.pptx
BPAC WITH UFSBI GENERAL PRESENTATION 18_05_2017-1.pptxBPAC WITH UFSBI GENERAL PRESENTATION 18_05_2017-1.pptx
BPAC WITH UFSBI GENERAL PRESENTATION 18_05_2017-1.pptxMohammedJunaid861692
 
CALL ON ➥8923113531 🔝Call Girls Chinhat Lucknow best sexual service Online
CALL ON ➥8923113531 🔝Call Girls Chinhat Lucknow best sexual service OnlineCALL ON ➥8923113531 🔝Call Girls Chinhat Lucknow best sexual service Online
CALL ON ➥8923113531 🔝Call Girls Chinhat Lucknow best sexual service Onlineanilsa9823
 
(PARI) Call Girls Wanowrie ( 7001035870 ) HI-Fi Pune Escorts Service
(PARI) Call Girls Wanowrie ( 7001035870 ) HI-Fi Pune Escorts Service(PARI) Call Girls Wanowrie ( 7001035870 ) HI-Fi Pune Escorts Service
(PARI) Call Girls Wanowrie ( 7001035870 ) HI-Fi Pune Escorts Serviceranjana rawat
 
April 2024 - Crypto Market Report's Analysis
April 2024 - Crypto Market Report's AnalysisApril 2024 - Crypto Market Report's Analysis
April 2024 - Crypto Market Report's Analysismanisha194592
 
{Pooja: 9892124323 } Call Girl in Mumbai | Jas Kaur Rate 4500 Free Hotel Del...
{Pooja:  9892124323 } Call Girl in Mumbai | Jas Kaur Rate 4500 Free Hotel Del...{Pooja:  9892124323 } Call Girl in Mumbai | Jas Kaur Rate 4500 Free Hotel Del...
{Pooja: 9892124323 } Call Girl in Mumbai | Jas Kaur Rate 4500 Free Hotel Del...Pooja Nehwal
 
Edukaciniai dropshipping via API with DroFx
Edukaciniai dropshipping via API with DroFxEdukaciniai dropshipping via API with DroFx
Edukaciniai dropshipping via API with DroFxolyaivanovalion
 
Best VIP Call Girls Noida Sector 39 Call Me: 8448380779
Best VIP Call Girls Noida Sector 39 Call Me: 8448380779Best VIP Call Girls Noida Sector 39 Call Me: 8448380779
Best VIP Call Girls Noida Sector 39 Call Me: 8448380779Delhi Call girls
 
Delhi Call Girls Punjabi Bagh 9711199171 ☎✔👌✔ Whatsapp Hard And Sexy Vip Call
Delhi Call Girls Punjabi Bagh 9711199171 ☎✔👌✔ Whatsapp Hard And Sexy Vip CallDelhi Call Girls Punjabi Bagh 9711199171 ☎✔👌✔ Whatsapp Hard And Sexy Vip Call
Delhi Call Girls Punjabi Bagh 9711199171 ☎✔👌✔ Whatsapp Hard And Sexy Vip Callshivangimorya083
 
Chintamani Call Girls: 🍓 7737669865 🍓 High Profile Model Escorts | Bangalore ...
Chintamani Call Girls: 🍓 7737669865 🍓 High Profile Model Escorts | Bangalore ...Chintamani Call Girls: 🍓 7737669865 🍓 High Profile Model Escorts | Bangalore ...
Chintamani Call Girls: 🍓 7737669865 🍓 High Profile Model Escorts | Bangalore ...amitlee9823
 
Midocean dropshipping via API with DroFx
Midocean dropshipping via API with DroFxMidocean dropshipping via API with DroFx
Midocean dropshipping via API with DroFxolyaivanovalion
 
Smarteg dropshipping via API with DroFx.pptx
Smarteg dropshipping via API with DroFx.pptxSmarteg dropshipping via API with DroFx.pptx
Smarteg dropshipping via API with DroFx.pptxolyaivanovalion
 
Call Girls 🫤 Dwarka ➡️ 9711199171 ➡️ Delhi 🫦 Two shot with one girl
Call Girls 🫤 Dwarka ➡️ 9711199171 ➡️ Delhi 🫦 Two shot with one girlCall Girls 🫤 Dwarka ➡️ 9711199171 ➡️ Delhi 🫦 Two shot with one girl
Call Girls 🫤 Dwarka ➡️ 9711199171 ➡️ Delhi 🫦 Two shot with one girlkumarajju5765
 
VIP Call Girls Service Miyapur Hyderabad Call +91-8250192130
VIP Call Girls Service Miyapur Hyderabad Call +91-8250192130VIP Call Girls Service Miyapur Hyderabad Call +91-8250192130
VIP Call Girls Service Miyapur Hyderabad Call +91-8250192130Suhani Kapoor
 
Generative AI on Enterprise Cloud with NiFi and Milvus
Generative AI on Enterprise Cloud with NiFi and MilvusGenerative AI on Enterprise Cloud with NiFi and Milvus
Generative AI on Enterprise Cloud with NiFi and MilvusTimothy Spann
 

Recently uploaded (20)

Halmar dropshipping via API with DroFx
Halmar  dropshipping  via API with DroFxHalmar  dropshipping  via API with DroFx
Halmar dropshipping via API with DroFx
 
Introduction-to-Machine-Learning (1).pptx
Introduction-to-Machine-Learning (1).pptxIntroduction-to-Machine-Learning (1).pptx
Introduction-to-Machine-Learning (1).pptx
 
Vip Model Call Girls (Delhi) Karol Bagh 9711199171✔️Body to body massage wit...
Vip Model  Call Girls (Delhi) Karol Bagh 9711199171✔️Body to body massage wit...Vip Model  Call Girls (Delhi) Karol Bagh 9711199171✔️Body to body massage wit...
Vip Model Call Girls (Delhi) Karol Bagh 9711199171✔️Body to body massage wit...
 
꧁❤ Greater Noida Call Girls Delhi ❤꧂ 9711199171 ☎️ Hard And Sexy Vip Call
꧁❤ Greater Noida Call Girls Delhi ❤꧂ 9711199171 ☎️ Hard And Sexy Vip Call꧁❤ Greater Noida Call Girls Delhi ❤꧂ 9711199171 ☎️ Hard And Sexy Vip Call
꧁❤ Greater Noida Call Girls Delhi ❤꧂ 9711199171 ☎️ Hard And Sexy Vip Call
 
BPAC WITH UFSBI GENERAL PRESENTATION 18_05_2017-1.pptx
BPAC WITH UFSBI GENERAL PRESENTATION 18_05_2017-1.pptxBPAC WITH UFSBI GENERAL PRESENTATION 18_05_2017-1.pptx
BPAC WITH UFSBI GENERAL PRESENTATION 18_05_2017-1.pptx
 
Delhi 99530 vip 56974 Genuine Escort Service Call Girls in Kishangarh
Delhi 99530 vip 56974 Genuine Escort Service Call Girls in  KishangarhDelhi 99530 vip 56974 Genuine Escort Service Call Girls in  Kishangarh
Delhi 99530 vip 56974 Genuine Escort Service Call Girls in Kishangarh
 
CALL ON ➥8923113531 🔝Call Girls Chinhat Lucknow best sexual service Online
CALL ON ➥8923113531 🔝Call Girls Chinhat Lucknow best sexual service OnlineCALL ON ➥8923113531 🔝Call Girls Chinhat Lucknow best sexual service Online
CALL ON ➥8923113531 🔝Call Girls Chinhat Lucknow best sexual service Online
 
(PARI) Call Girls Wanowrie ( 7001035870 ) HI-Fi Pune Escorts Service
(PARI) Call Girls Wanowrie ( 7001035870 ) HI-Fi Pune Escorts Service(PARI) Call Girls Wanowrie ( 7001035870 ) HI-Fi Pune Escorts Service
(PARI) Call Girls Wanowrie ( 7001035870 ) HI-Fi Pune Escorts Service
 
April 2024 - Crypto Market Report's Analysis
April 2024 - Crypto Market Report's AnalysisApril 2024 - Crypto Market Report's Analysis
April 2024 - Crypto Market Report's Analysis
 
{Pooja: 9892124323 } Call Girl in Mumbai | Jas Kaur Rate 4500 Free Hotel Del...
{Pooja:  9892124323 } Call Girl in Mumbai | Jas Kaur Rate 4500 Free Hotel Del...{Pooja:  9892124323 } Call Girl in Mumbai | Jas Kaur Rate 4500 Free Hotel Del...
{Pooja: 9892124323 } Call Girl in Mumbai | Jas Kaur Rate 4500 Free Hotel Del...
 
Edukaciniai dropshipping via API with DroFx
Edukaciniai dropshipping via API with DroFxEdukaciniai dropshipping via API with DroFx
Edukaciniai dropshipping via API with DroFx
 
Best VIP Call Girls Noida Sector 39 Call Me: 8448380779
Best VIP Call Girls Noida Sector 39 Call Me: 8448380779Best VIP Call Girls Noida Sector 39 Call Me: 8448380779
Best VIP Call Girls Noida Sector 39 Call Me: 8448380779
 
Delhi Call Girls Punjabi Bagh 9711199171 ☎✔👌✔ Whatsapp Hard And Sexy Vip Call
Delhi Call Girls Punjabi Bagh 9711199171 ☎✔👌✔ Whatsapp Hard And Sexy Vip CallDelhi Call Girls Punjabi Bagh 9711199171 ☎✔👌✔ Whatsapp Hard And Sexy Vip Call
Delhi Call Girls Punjabi Bagh 9711199171 ☎✔👌✔ Whatsapp Hard And Sexy Vip Call
 
Chintamani Call Girls: 🍓 7737669865 🍓 High Profile Model Escorts | Bangalore ...
Chintamani Call Girls: 🍓 7737669865 🍓 High Profile Model Escorts | Bangalore ...Chintamani Call Girls: 🍓 7737669865 🍓 High Profile Model Escorts | Bangalore ...
Chintamani Call Girls: 🍓 7737669865 🍓 High Profile Model Escorts | Bangalore ...
 
Midocean dropshipping via API with DroFx
Midocean dropshipping via API with DroFxMidocean dropshipping via API with DroFx
Midocean dropshipping via API with DroFx
 
꧁❤ Aerocity Call Girls Service Aerocity Delhi ❤꧂ 9999965857 ☎️ Hard And Sexy ...
꧁❤ Aerocity Call Girls Service Aerocity Delhi ❤꧂ 9999965857 ☎️ Hard And Sexy ...꧁❤ Aerocity Call Girls Service Aerocity Delhi ❤꧂ 9999965857 ☎️ Hard And Sexy ...
꧁❤ Aerocity Call Girls Service Aerocity Delhi ❤꧂ 9999965857 ☎️ Hard And Sexy ...
 
Smarteg dropshipping via API with DroFx.pptx
Smarteg dropshipping via API with DroFx.pptxSmarteg dropshipping via API with DroFx.pptx
Smarteg dropshipping via API with DroFx.pptx
 
Call Girls 🫤 Dwarka ➡️ 9711199171 ➡️ Delhi 🫦 Two shot with one girl
Call Girls 🫤 Dwarka ➡️ 9711199171 ➡️ Delhi 🫦 Two shot with one girlCall Girls 🫤 Dwarka ➡️ 9711199171 ➡️ Delhi 🫦 Two shot with one girl
Call Girls 🫤 Dwarka ➡️ 9711199171 ➡️ Delhi 🫦 Two shot with one girl
 
VIP Call Girls Service Miyapur Hyderabad Call +91-8250192130
VIP Call Girls Service Miyapur Hyderabad Call +91-8250192130VIP Call Girls Service Miyapur Hyderabad Call +91-8250192130
VIP Call Girls Service Miyapur Hyderabad Call +91-8250192130
 
Generative AI on Enterprise Cloud with NiFi and Milvus
Generative AI on Enterprise Cloud with NiFi and MilvusGenerative AI on Enterprise Cloud with NiFi and Milvus
Generative AI on Enterprise Cloud with NiFi and Milvus
 

Decode Player Feedback Using Sentiment and Topic Analysis

  • 1. Nikola Vasiljević – Head Of Insights & Analytics at Mad Head Games LEVERAGING SENTIMENT AND TOPIC DETECTION FOR DECODING PLAYER FEEDBACK
  • 2. THE AIM IS TO GET YOU FAMILIAR WITH SENTIMENT AND TOPIC ANALYSIS AND INSPIRE YOU TO USE THEM (for free) IN YOUR DECISION-MAKING STACK. YOU WILL LEARN: 1. INTRO TO SENTIMENT AND TOPIC 2. HOW WE UTILIZE THEM, WITH EXAMPLES 3. CHALLENGES WE FACED 4. WHAT METHODOLOGIES AND ALGORITHM WE USE 5. KEY TAKEAWAYS WITH BENEFITS
  • 3. INTRO TO SENTIMENT AND TOPIC ANALYSIS: ANALYZE YOUR PLAYERS’ REVIEW AND COMMENTS SENTIMENT ANALYSIS An area of qualitative research and NLP that analyzes textual input and assigns sentiment: • Positive • Neutral • Negative Common application: Analysis of customer satisfaction Can be expanded to detect: • More sentiment detail – mixed, very positive, very negative • Specific emotions (happy, sad, thrilled, excited, angry…) • Priority/Urgency • Intentions (interested, not interested, purchase intent…) QUALITATIVE DATA ANALYSIS - TOPIC DETECTION ANALYSIS • A set of research and data analysis methods aimed at systematically interpreting and extracting meaningful insights from non-numerical data • Common application: Contextual analysis of consumer feedback 8.5
  • 4. WE USE IT TO: ANALYZE PLAYERS’ COMMENTS AND REVIEWS ON THE INTERNET, AND LEARN WHAT THEY ARE TALKING ABOUT AND HOW THEY FEEL ABOUT IT This enables us to deliver insights on how we can make and improve games based on players’ feedback • Use main genre drivers as pillars • Make different game design decision • Solve bugs • Balance out our games • Refine our communication to manage players’ expectations • Patch up our games USER ATTITUDE AND SATISFACTION ANALYZE COMPETITION PLAYER SEGMENTATION ANALYZE GENRES’ MAIN DRIVERS
  • 5. CHALLENGES: PROVIDE INSIGHTFUL AND TIMELY FEEDBACK DURING THE STEPS OF THE PRODUCTION CYCLE TRADITIONAL QUALITATIVE METHODS • Require unbiased experts • Take a lot of time • Hard to scale for big data • Cannot be fully automated ML – NLP and LLMs • Understanding irony, double meaning, sarcasm • Broken language and mistakes • Industry-specific slang – n00b, OP, take my money… • Understanding context (“Andromeda anyone?“) To get useful, actionable insights, it requires: • a mixture of methods • a lot of fine tuning • a human “bus driver“
  • 6. • CHALLENGING, IN A POSITIVE WAY, AND FUN • ELEMENTAL WEAPONS AND ENVIRONMENTAL DAMAGE SYSTEM IS COOL • BOSS FIGHTS ARE EXCITING • LEARNING ATTACK PATTERNS IS REWARDING • REQUIRES SKILL, WHICH IS REWARDING • UNFAIR AND IMBALANCED COMBAT • BOSS DIFFICULTY, ESPECIALLY THE FIRST ONE IS UNFAIR • NO PILLAR TO SAVE THE GAME BEFORE THE BOSS FIGHT • WAS THERE FOR THE STORY, BUT THE COMBAT IS TOO MUCH • BOSS DIFFICULTY SPIKE • WHY IS IT SOULSIKE WHAT NEXT? EXAMPLE FROM SCARS ABOVE: A TOP DOWN ANALYSIS OF PLAYERS’ SENTIMENT AND TOPIC DETECTION Sub-topic Sub-topic
  • 7. GROUNDED THEORY: FORM THEORIES AND HYPOTHESES BASED ON DATA, TEST THEM AND TAKE ACTION What we did • Conducted additional research to find out the nature and the differences between story-driven players and combat-driven players, which gave us confirmation for our hypothesis • Lowered overall difficulty • Adjusted the design of some boss fights • Reworked the difficulty settings by making game design changes to the EASY mode  removed some souslike elements, like respawning enemies ADJUSTING COMBAT / BOSS DIFFICULTY BY MAKING IT EASIER WILL MAKE THE GAME PLAYABLE FOR NARRATIVE-FIRST PLAYERS WHILE NOT HARMING THE EXPERIENCE OF THE SOULSLIKE PLAYERS TARGET SUB-GROUP Why is it important to develop theories? • Game design elements are intertwined and work in balance – tweaking one will inevitably impact others, so taking on them separately will get you nowhere • We have multiple target sub-groups which sometimes are driven by totally opposite drivers  too much catering to one may repel another
  • 8. SENTIMENT ANALYSIS: A MIX OF APPROACHES
  • 9. SENTIMENT ANALYSIS: A MIX OF APPROACHES 1. Devise a sample (N=1000) 2. Go manually through each review/comment and assign sentiment 3. Use it as a training model
  • 10. SENTIMENT ANALYSIS: A MIX OF APPROACHES 1. NLTK Python library 2. Calculate sentiment for each review/comment 3. We used a pre-trained model with fine tuning
  • 11. Use OpenAI API to make your own assistant which will do sentiment analysis for you SENTIMENT ANALYSIS: A MIX OF APPROACHES
  • 12. NLTK OVERVIEW: A PRE-TRAINED MODEL WITH FINE TUNING WORKS SOLIDLY WELL 1. Text preprocessing • Remove stop-words (words of little value (the, and, is, in, of…) • Word exclusion – if there are any particular words to exclude • Tokenization – the text is transformed into individual tokens to facilitate analysis at a more granular level • Stemming – removing suffixes to get the root of the word (run, running, runs run) • Lemmatization - reducing words to their base or dictionary form, known as the "lemma." It considers the meaning of the word and applies morphological analysis to obtain the root form; good, better, best  good 2. Sentiment intensity analyzer • Apply sentiment scores – calculates sentiment for textual input based on tokens’ sentiment scores 3. Fine-tuning • Manually adjusting sentiment thresholds • Adding a custom dictionary (“sick, OP, take my money…“) Comment Sentiment 0 Holy hell. That was a movie-level trailer!<br>... Positive 1 Animations are so bad seeing the graphics it d... Negative 2 Pyramids and darkness? Destiny is leaking Neutral 3 <b>facepalm</b> the running animation is still... Neutral 4 This looks interesting Positive .. ... ... 111 oh again STONG women? facepalm Negative 112 Alles sehr schön. Aber zuerst zusammen die Num... Positive 113 Let&#39;s goooo Positive 114 Awesome: I&#39;m eager for this game Positive 115 Really looking forward to this game! Positive THE OUTPUT ASSIGNING SENTIMENT SCORES
  • 13. OPEN AI: PROMPT-ENGINEER YOUR OWN ASSISTANT 1. Specify model (gpt 3.5 turbo, gpt-4…) 2. Define messages: 1. Role of the messenger: system, user, assistant 2. Content of the message – tell it what it should do Prompt engineering step-by-step*: a) system message b) very specific instructions, each given in a separate prompt c) Provide input data d) Add contextual information e) Set temperature (controls randomness of the output) f) Specify desired output format (json, csv) * Goran Milovanović: Generative AI in Action: Three case studies with OpenAI GPT(3-3.5) THE OUTPUT DEFINE YOUR ASSISTANT
  • 14. TOPIC ANALYSIS: A MIX OF APPROACHES
  • 15. TOPIC ANALYSIS: A MIX OF APPROACHES 1. Devise a sample (N=1000) 2. Go manually through each review/comment and assign tags 3. Synthesis – group tags into subtopics and topics
  • 16. TOPIC ANALYSIS: A MIX OF APPROACHES 1. Take the keywords devised with qualitative coding 2. Setup IF THEN ELSE classification rules for keyword search of your text input, and group by topic and subtopic
  • 17. TOPIC ANALYSIS: A MIX OF APPROACHES BERT AND LDA on their own didn’t get us to where we wanted. LDA + OPEN AI LLM: 1. Preprocess the data (as in NLTK) 2. Extract topics and associeated words with LDA to a list 3. Take the extracted topics and generate a more user-friendly output
  • 18. TOPIC ANALYSIS: A MIX OF APPROACHES Use OpenAI API to make your own assistant which will do sentiment analysis for you
  • 19. QUALITATIVE CODING AND ASPECT BASED TOPIC DETECTION: MANUAL WORK WITH HIGH REWARD COMBAT DIFFICULTY HITTING WEAKSPOTS FIRE WEAPON RELOAD SHORT WEAKSPOT OPPORTUNITY WINDOW ..It was sooo unfair hitting the glowing part of the chest with the fire weapon, as reload times are too long so you miss the opporutnity window… COMBAT DIFFICULTY WEAPONS COMBAT DESIGN COMBAT DIFFICULTY SHOOTING RELOAD COMBAT DESIGN 1. Take the keywords devised with qualitative coding 2. Setup IF THEN ELSE or REGEX classification rules for keyword search of text input, and group by topic&subtopic ASPECT BASED HARD-CODED TOPIC DETECTION
  • 20. LDA + OPEN AI LLM: A WELL-ROUNDED IMPRESSIONS SUMMARY… 1. Define a function (A) to extract topics and associated keywords from the input 2. Provide it parameters 1. Input parameters 2. Number of topics to discover 3. Number of words to include per topic 4. Output parameters – a list of topics with sublists of keywords 3. Define a function (B) to take the previous output and generate descriptive texts about detected topics using detected keywords 4. Define parameters 1. LLM 2. Input parameters 3. Number of topics 4. Number of words per topic 5. Output format OUTPUT FUNCTION A FUNCTION B (a) (b)
  • 21. OPEN AI TOPIC DETECTION: UNDERSTANDS CONTEXT BUT LIKES TO TALK IN SYNONYMS AND IS OCCASIONALLY STUBBORN 1. Specify model (gpt 3.5 turbo, gpt-4…) 2. Define messages: 1. Role of the messenger: system, user, assistant 2. Content of the message – tell it what it should do Prompt engineering step-by-step*: a) system message b) very specific instructions, each given in a separate prompt c) Provide input data d) Add contextual information e) Set temperature (controls randomness of the output) f) Specify desired output format (json, csv) DEFINE YOUR ASSISTANT OUTPUT Comment Sentiment Topic 1 Topic 2 Topic 3 Topic 4 Topic 5 0 Holy hell. That was a movie-level trailer!<br>... Positive Appreciation Trailer Comparison Quality 1 Animations are so bad seeing the graphics it d... Negative Animation Graphics Quality 2 Pyramids and darkness? Destiny is leaking Neutral Comparison Darkness 3 <b>facepalm</b> the running animation is still... Mixed Animation Appreciation 4 This looks interesting Positive Interest Appearance .. ... ... 111 oh again STONG women? facepalm Negative Character Face 112 Alles sehr schön. Aber zuerst zusammen die Num... Mixed Potential 113 Let&#39;s goooo Positive Movement 114 Awesome: I&#39;m eager for this game Positive Interest 115 Really looking forward to this game! Positive Anticipation [116 rows x 2 columns]
  • 22. A HOLISTIC APPROACH TO ANALYZING PLAYERS’S FEEDBACK SENTIMENT AND TOPIC: HOW WE DO IT INITIAL MANUAL SENTIMENT AND TOPIC ANALYSIS USING THE QUALITATIVE CODING METHODOLOGY • Devise a N=1000 sample, and perform a manual sentiment and topic analysis with qualitative coding • Use it as a training dataset for ML • Listed topics, subtopics and keywords are used as input instructions for aspect-based topic analysis, as well as for ML fine tuning • You get to really acquaint yourself with the data and get to know feeling, the players, the persons behind them  you get into players’ shoes ASPECT BASED TOPIC ANALYSIS: • ML/AI Sentiment analysis (fine tuned using the aspects from previous step) • Use the aspects and keywords to form keywordssubtopicstopics and fine tune your tool/algorithm to look for them (you don’t need ML for this, a simple IF THEN ELSE to count topics and subtopics mentioned based on keywords will do (we do this in Tableau) • It’s all-encompassing LLM AND OPEN AI MODELS • Compare results • Validate your findings from aspect-based models • Enhance the previous results – something may pop up that you didn’t encompass with your manually defined aspects • Keep fine tuning GROUNDED THEORY  CONNECTING THE DOTS BY DEVELOPING INSIGHTS AND HYPOTHESES BASED ON DATA • It aims to develop theories or concepts grounded in the data, allowing patterns and themes to emerge organically from the data without preconceived notions.
  • 23. YOU CAN BUILD YOUR OWN AUTOMATED TOOLS: THIS IS OUR COMPETITION ANALYSIS TOOL YouTube Scrapping script NLTK (fine- tuned) Aspect-based
  • 24. KEY TAKEAWAYS USE SENTIMENT AND TOPIC FOR PLAYER FEEDBACK ANALYSIS • There’s an abundance of free data • You will learn a lot about your players • It WILL help you make your games better USE THE DATA TO GET TO KNOW YOUR PLAYERS • Understand your players and develop segments • Understand that you have many target subgroups • Understand what are the main drivers for each of them • But, also understand you can’t please everyone DEVELOP YOUR OWN FRAMEWORK • Experiment with all methodologies, don’t discriminate • Traditional (qualitative coding) • Automated (ML algorithms, OpenAI) • Combine them and find out what works for you DEVELOP INSIGHTS USING GROUNDED THEORY MAKE IT A PART OF YOUR DECISION- MAKING STACK DO THE MANUAL WORK • Take the time and do an initial manual analysis on N=1000 reviews or comments • It will familiarize you with the data • Use it as a training set • Go beyond plain data and develop theories with your game designer, based on the data and knowledge you acquired – not just with Sentiment and Topic analysis, but with all of your insights • Test the hypotheses Analyze your game life-cycle and find out how and when it can be used in to make decisions and connect with your decision makers: • During production • During soft-launch / demo • When a game is live