Keynote delivered at the 2013 Workshop on
Using Predictive Coding in E-Discovery (DESI V), about minimizing the cost of human review following an automated classification pass
Oleksander Krakovetskyi "Explaining a Machine Learning blackbox"Fwdays
As Data Scientists we want to understand machine learning models we have built. “Why did my model make this mistake?”, “Does my model discriminate?”, “How can I understand and trust the model's decisions?”, “Does my model satisfy legal requirements?” are commonly asked questions.
In this presentation we will talk about machine learning explainability and interpretability - two concepts that could help us really understand ML models.
Website: https://fwdays.com/en/event/data-science-fwdays-2019/review/explaining-a-machine-learning-blackbox
What is pattern recognition (lecture 4 of 6)Randa Elanwar
In this series I intend to simplify a beautiful branch of computer science that we as humans use it in everyday life without knowing. Pattern recognition is a sub-branch of the computer vision research and is tightly related to digital signal processing research as well as machine learning and artificial intelligence.
Keynote delivered at the 2013 Workshop on
Using Predictive Coding in E-Discovery (DESI V), about minimizing the cost of human review following an automated classification pass
Oleksander Krakovetskyi "Explaining a Machine Learning blackbox"Fwdays
As Data Scientists we want to understand machine learning models we have built. “Why did my model make this mistake?”, “Does my model discriminate?”, “How can I understand and trust the model's decisions?”, “Does my model satisfy legal requirements?” are commonly asked questions.
In this presentation we will talk about machine learning explainability and interpretability - two concepts that could help us really understand ML models.
Website: https://fwdays.com/en/event/data-science-fwdays-2019/review/explaining-a-machine-learning-blackbox
What is pattern recognition (lecture 4 of 6)Randa Elanwar
In this series I intend to simplify a beautiful branch of computer science that we as humans use it in everyday life without knowing. Pattern recognition is a sub-branch of the computer vision research and is tightly related to digital signal processing research as well as machine learning and artificial intelligence.
Secuencia didáctica para el aprendizaje del adjetivo en el grado quinto de básica primaría del Colegio Universitario Socorro Santander- Sede D Cooperativo.
Apresentação de Negócios Welions - Como Funciona a Welions igual TelexfreeO Melhor do MMN
Apresentação de Negócios Welions - Como Funciona a Welions igual Telexfree
Saiba mais: http://www.omelhordommn.com.br/welions
Formas de Ganhos:
Bônus Publicidade
Vendas Diretas
Residual de Clientes
Bônus Binário
Bônus Trinário
Residual de Recompra
Master Club
Automated Classification and Quantification of Verbatims via Machine...Fabrizio Sebastiani
Keynote delivered at the 2013 Conference of the Association for Survey Computing, about automatically classifying open-ended answers and about quantifying their distribution across the codes of interest
“Towards Multi-Step Expert Advice for Cognitive Computing” - Dr. Achim Rettin...diannepatricia
Dr. Achim Rettinger from Karlsruhe Institute of Technology presented this today as part of the Cognitive Systems Institute Speaker Series on October 13, 2016
Daniel Samaan: ChatGPT and the Future of WorkEdunomica
Daniel Samaan: ChatGPT and the Future of Work
People Analytics Conference 2023 Summer
Website: https://pacamp.org
Youtube: https://www.youtube.com/channel/UCeHtPZ_ZLZ-nHFMUCXY81RQ
FB: https://www.facebook.com/pacamporg
Intelligent Hiring with Resume Parser and Ranking using Natural Language Proc...Zainul Sayed
Using Natural Language Processing(NLP) and (ML)Machine Learning to rank the resumes according to the given constraint, this intelligent system ranks the resume of any format according to the given constraints or following the requirements provided by the client company. We will basically take the bulk of input resume from the client company and that client company will also provided the requirement and the constraints according to which the resume shall be ranked by our system. Moreover the details acquired from the resumes, our system shall be reading the candidates social profiles (like LinkedIn, Github etc) which will the more genuine information about that candidate.
[MMIR@MM2023] On Popularity Bias of Multimodal-aware Recommender Systems: A M...Daniele Malitesta
Slides for the paper "On Popularity Bias of Multimodal-aware Recommender Systems: A Modalities-driven Analysis", accepted and presented at the 1st International Workshop on Deep Multimodal Learning for Information Retrieval, co-located with the 31st ACM International Conference on Multimedia (MMIR@MM'23).
Paper: https://dl.acm.org/doi/abs/10.1145/3606040.3617441
Code: https://github.com/sisinflab/MultiMod-Popularity-Bias
Performance Analysis of Supervised Machine Learning Techniques for Sentiment ...Biswaranjan Samal
Wide use of internet and web applications like, feedback collection systems are now making peoples smarter. In these applications, peoples used to give their feedback about the movies, products, services, etc through which they have gone, and this feedback are publicly available for future references. It is a tedious task for the machines to identify the feedback types, i:e positive or negative. And here Machine Learning Techniques plays vital roles to train the machine and make it intelligent so that the machine will be able to identify the feedback type which may give more benefits and features for those web applications and the users. There are many supervised machine learning techniques are available so it is a difficult task to choose the best one. In this paper, we have collected the movie review datasets of different sizes and have selected some of the widely used and popular supervised machine learning algorithms, for training the model. So that the model will be able to categorize the review. Python's NLTK package along with the WinPython and Spyder are used for processing the movie reviews. Then Python's sklearn package is used for training the model and finding the accuracy of the model.
For the full video of this presentation, please visit:
https://www.embedded-vision.com/platinum-members/embedded-vision-alliance/embedded-vision-training/videos/pages/sept-2017-alliance-vitf-samek
For more information about embedded vision, please visit:
http://www.embedded-vision.com
Dr. Wojciech Samek of the Fraunhofer Heinrich Hertz Institute delivers the presentation "Methods for Understanding How Deep Neural Networks Work" at the Embedded Vision Alliance's September 2017 Vision Industry and Technology Forum. In his presentation, Dr. Samek covers the following topics:
▪ Unbeatable AI systems
▪ Deep neural network overview
▪ Opening the "black box"
▪ Summary
Secuencia didáctica para el aprendizaje del adjetivo en el grado quinto de básica primaría del Colegio Universitario Socorro Santander- Sede D Cooperativo.
Apresentação de Negócios Welions - Como Funciona a Welions igual TelexfreeO Melhor do MMN
Apresentação de Negócios Welions - Como Funciona a Welions igual Telexfree
Saiba mais: http://www.omelhordommn.com.br/welions
Formas de Ganhos:
Bônus Publicidade
Vendas Diretas
Residual de Clientes
Bônus Binário
Bônus Trinário
Residual de Recompra
Master Club
Automated Classification and Quantification of Verbatims via Machine...Fabrizio Sebastiani
Keynote delivered at the 2013 Conference of the Association for Survey Computing, about automatically classifying open-ended answers and about quantifying their distribution across the codes of interest
“Towards Multi-Step Expert Advice for Cognitive Computing” - Dr. Achim Rettin...diannepatricia
Dr. Achim Rettinger from Karlsruhe Institute of Technology presented this today as part of the Cognitive Systems Institute Speaker Series on October 13, 2016
Daniel Samaan: ChatGPT and the Future of WorkEdunomica
Daniel Samaan: ChatGPT and the Future of Work
People Analytics Conference 2023 Summer
Website: https://pacamp.org
Youtube: https://www.youtube.com/channel/UCeHtPZ_ZLZ-nHFMUCXY81RQ
FB: https://www.facebook.com/pacamporg
Intelligent Hiring with Resume Parser and Ranking using Natural Language Proc...Zainul Sayed
Using Natural Language Processing(NLP) and (ML)Machine Learning to rank the resumes according to the given constraint, this intelligent system ranks the resume of any format according to the given constraints or following the requirements provided by the client company. We will basically take the bulk of input resume from the client company and that client company will also provided the requirement and the constraints according to which the resume shall be ranked by our system. Moreover the details acquired from the resumes, our system shall be reading the candidates social profiles (like LinkedIn, Github etc) which will the more genuine information about that candidate.
[MMIR@MM2023] On Popularity Bias of Multimodal-aware Recommender Systems: A M...Daniele Malitesta
Slides for the paper "On Popularity Bias of Multimodal-aware Recommender Systems: A Modalities-driven Analysis", accepted and presented at the 1st International Workshop on Deep Multimodal Learning for Information Retrieval, co-located with the 31st ACM International Conference on Multimedia (MMIR@MM'23).
Paper: https://dl.acm.org/doi/abs/10.1145/3606040.3617441
Code: https://github.com/sisinflab/MultiMod-Popularity-Bias
Performance Analysis of Supervised Machine Learning Techniques for Sentiment ...Biswaranjan Samal
Wide use of internet and web applications like, feedback collection systems are now making peoples smarter. In these applications, peoples used to give their feedback about the movies, products, services, etc through which they have gone, and this feedback are publicly available for future references. It is a tedious task for the machines to identify the feedback types, i:e positive or negative. And here Machine Learning Techniques plays vital roles to train the machine and make it intelligent so that the machine will be able to identify the feedback type which may give more benefits and features for those web applications and the users. There are many supervised machine learning techniques are available so it is a difficult task to choose the best one. In this paper, we have collected the movie review datasets of different sizes and have selected some of the widely used and popular supervised machine learning algorithms, for training the model. So that the model will be able to categorize the review. Python's NLTK package along with the WinPython and Spyder are used for processing the movie reviews. Then Python's sklearn package is used for training the model and finding the accuracy of the model.
For the full video of this presentation, please visit:
https://www.embedded-vision.com/platinum-members/embedded-vision-alliance/embedded-vision-training/videos/pages/sept-2017-alliance-vitf-samek
For more information about embedded vision, please visit:
http://www.embedded-vision.com
Dr. Wojciech Samek of the Fraunhofer Heinrich Hertz Institute delivers the presentation "Methods for Understanding How Deep Neural Networks Work" at the Embedded Vision Alliance's September 2017 Vision Industry and Technology Forum. In his presentation, Dr. Samek covers the following topics:
▪ Unbeatable AI systems
▪ Deep neural network overview
▪ Opening the "black box"
▪ Summary
TEXT SENTIMENTS FOR FORUMS HOTSPOT DETECTIONijistjournal
The user generated content on the web grows rapidly in this emergent information age. The evolutionary changes in technology make use of such information to capture only the user’s essence and finally the useful information are exposed to information seekers. Most of the existing research on text information processing, focuses in the factual domain rather than the opinion domain. In this paper we detect online hotspot forums by computing sentiment analysis for text data available in each forum. This approach analyses the forum text data and computes value for each word of text. The proposed approach combines K-means clustering and Support Vector Machine with PSO (SVM-PSO) classification algorithm that can be used to group the forums into two clusters forming hotspot forums and non-hotspot forums within the current time span. The proposed system accuracy is compared with the other classification algorithms such as Naïve Bayes, Decision tree and SVM. The experiment helps to identify that K-means and SVM-PSO together achieve highly consistent results.
TEXT SENTIMENTS FOR FORUMS HOTSPOT DETECTIONijistjournal
The user generated content on the web grows rapidly in this emergent information age. The evolutionary changes in technology make use of such information to capture only the user’s essence and finally the useful information are exposed to information seekers. Most of the existing research on text information processing, focuses in the factual domain rather than the opinion domain. In this paper we detect online hotspot forums by computing sentiment analysis for text data available in each forum. This approach analyses the forum text data and computes value for each word of text. The proposed approach combines K-means clustering and Support Vector Machine with PSO (SVM-PSO) classification algorithm that can be used to group the forums into two clusters forming hotspot forums and non-hotspot forums within the current time span. The proposed system accuracy is compared with the other classification algorithms such as Naïve Bayes, Decision tree and SVM. The experiment helps to identify that K-means and SVM-PSO together achieve highly consistent results.
Self-organisation of Knowledge in Socio-technical Systems: A Coordination Per...Andrea Omicini
Some of the most peculiar traits of socio-technical systems (STS) in knowledge-intensive environments (KIE) – such as unpredictability of agents’ behaviour, ever-growing amount of information to manage, fast-paced production/consumption – tangle coordination of agents as well as coordination of information, by affecting, e.g., reachability by knowledge prosumers and manageability by the IT infrastructure. In this seminar we describe a novel approach to coordination of STS in KIE, grounded on the MoK (Molecules of Knowledge) model for knowledge self-organisation, and inspired to key concepts from the cognitive theory of BIC (behavioural implicit communication).
Jonas Schneider, Head of Engineering for Robotics, OpenAIMLconf
Machine Learning Systems at Scale:
OpenAI is a non-profit research company, discovering and enacting the path to safe artificial general intelligence. As part of our work, we regularly push the limits of scalability in cutting-edge ML algorithms. We’ve found that in many cases, designing the systems we build around the core algorithms is as important as designing the algorithms themselves. This means that many systems engineering areas, such as distributed computing, networking, and orchestration, are crucial for machine learning to succeed on large problems requiring thousands of computers. As a result, at OpenAI engineers and researchers work closely together to build these large systems as opposed to a strict researcher/engineer split. In this talk, we will go over some of the lessons we’ve learned, and how they come together in the design and internals of our system for learning-based robotics research.
Bio: Jonas leads technology development for OpenAI’s robotics group, developing methods to apply machine learning and AI to robots. He also helped build the infrastructure to scale OpenAI’s distributed ML systems to thousands of machines.
STATATHON: Unleashing the Power of Statistics in a 48-Hour Knowledge Extravag...sameer shah
"Join us for STATATHON, a dynamic 2-day event dedicated to exploring statistical knowledge and its real-world applications. From theory to practice, participants engage in intensive learning sessions, workshops, and challenges, fostering a deeper understanding of statistical methodologies and their significance in various fields."
Enhanced Enterprise Intelligence with your personal AI Data Copilot.pdfGetInData
Recently we have observed the rise of open-source Large Language Models (LLMs) that are community-driven or developed by the AI market leaders, such as Meta (Llama3), Databricks (DBRX) and Snowflake (Arctic). On the other hand, there is a growth in interest in specialized, carefully fine-tuned yet relatively small models that can efficiently assist programmers in day-to-day tasks. Finally, Retrieval-Augmented Generation (RAG) architectures have gained a lot of traction as the preferred approach for LLMs context and prompt augmentation for building conversational SQL data copilots, code copilots and chatbots.
In this presentation, we will show how we built upon these three concepts a robust Data Copilot that can help to democratize access to company data assets and boost performance of everyone working with data platforms.
Why do we need yet another (open-source ) Copilot?
How can we build one?
Architecture and evaluation
Learn SQL from basic queries to Advance queriesmanishkhaire30
Dive into the world of data analysis with our comprehensive guide on mastering SQL! This presentation offers a practical approach to learning SQL, focusing on real-world applications and hands-on practice. Whether you're a beginner or looking to sharpen your skills, this guide provides the tools you need to extract, analyze, and interpret data effectively.
Key Highlights:
Foundations of SQL: Understand the basics of SQL, including data retrieval, filtering, and aggregation.
Advanced Queries: Learn to craft complex queries to uncover deep insights from your data.
Data Trends and Patterns: Discover how to identify and interpret trends and patterns in your datasets.
Practical Examples: Follow step-by-step examples to apply SQL techniques in real-world scenarios.
Actionable Insights: Gain the skills to derive actionable insights that drive informed decision-making.
Join us on this journey to enhance your data analysis capabilities and unlock the full potential of SQL. Perfect for data enthusiasts, analysts, and anyone eager to harness the power of data!
#DataAnalysis #SQL #LearningSQL #DataInsights #DataScience #Analytics
Levelwise PageRank with Loop-Based Dead End Handling Strategy : SHORT REPORT ...Subhajit Sahu
Abstract — Levelwise PageRank is an alternative method of PageRank computation which decomposes the input graph into a directed acyclic block-graph of strongly connected components, and processes them in topological order, one level at a time. This enables calculation for ranks in a distributed fashion without per-iteration communication, unlike the standard method where all vertices are processed in each iteration. It however comes with a precondition of the absence of dead ends in the input graph. Here, the native non-distributed performance of Levelwise PageRank was compared against Monolithic PageRank on a CPU as well as a GPU. To ensure a fair comparison, Monolithic PageRank was also performed on a graph where vertices were split by components. Results indicate that Levelwise PageRank is about as fast as Monolithic PageRank on the CPU, but quite a bit slower on the GPU. Slowdown on the GPU is likely caused by a large submission of small workloads, and expected to be non-issue when the computation is performed on massive graphs.
06-04-2024 - NYC Tech Week - Discussion on Vector Databases, Unstructured Data and AI
Discussion on Vector Databases, Unstructured Data and AI
https://www.meetup.com/unstructured-data-meetup-new-york/
This meetup is for people working in unstructured data. Speakers will come present about related topics such as vector databases, LLMs, and managing data at scale. The intended audience of this group includes roles like machine learning engineers, data scientists, data engineers, software engineers, and PMs.This meetup was formerly Milvus Meetup, and is sponsored by Zilliz maintainers of Milvus.
Adjusting primitives for graph : SHORT REPORT / NOTESSubhajit Sahu
Graph algorithms, like PageRank Compressed Sparse Row (CSR) is an adjacency-list based graph representation that is
Multiply with different modes (map)
1. Performance of sequential execution based vs OpenMP based vector multiply.
2. Comparing various launch configs for CUDA based vector multiply.
Sum with different storage types (reduce)
1. Performance of vector element sum using float vs bfloat16 as the storage type.
Sum with different modes (reduce)
1. Performance of sequential execution based vs OpenMP based vector element sum.
2. Performance of memcpy vs in-place based CUDA based vector element sum.
3. Comparing various launch configs for CUDA based vector element sum (memcpy).
4. Comparing various launch configs for CUDA based vector element sum (in-place).
Sum with in-place strategies of CUDA mode (reduce)
1. Comparing various launch configs for CUDA based vector element sum (in-place).
Machine Learning and Automatic Text Classification: What's Next?
1. Machine Learning and Automatic Text Classification
What’s Next?
Fabrizio Sebastiani
(Joint work with Giacomo Berardi and Andrea Esuli)
Istituto di Scienza e Tecnologie dell’Informazione
Consiglio Nazionale delle Ricerche
56124 Pisa, Italy
ASC Methods Conference
Winchester, UK – September 6-7, 2013
2. Prequel: ML for Automated Verbatim Coding
In the last 10 years we have championed an approach to automatically coding
open-ended answers (“verbatims”) based on “machine learning”;
2003 : D. Giorgetti, I. Prodanof, and F. Sebastiani. Automatic Coding of
Open-ended Questions Using Text Categorization Techniques. Proceedings of
the 4th International Conference of the Association for Survey Computing,
Warwick, UK, pp. 173-–184.
2007 : T. Macer, M. Pearson, and F. Sebastiani. Cracking the Code: What
Customers Say, in Their Own Words. In Proceedings of the 50th Annual
Conference of the Market Research Society, Brighton, UK. (Best New
Thinking Award, Shortlisted for Best Paper Award and for ASC/MRS Tech
Effectiveness Award)
2010 : A. Esuli and F. Sebastiani. Machines that learn how to code
open-ended survey data. International Journal of Market Research, 52(6).
(Shortlisted for best 2010 IJMR paper)
Fabrizio Sebastiani (ISTI-CNR, Pisa, IT) ML & ATC: What’s Next? ASC Methods Conference 2 / 41
3. Prequel: ML for Automated Verbatim Coding (cont’d)
Based on these principles we have built a software system, called VCS
(“Verbatim Coding System”), which has been variously applied to coding
surveys in the social sciences, customer relationship management, and market
research.
VCS is based on a “supervised learning” metaphor : the classifier learns (or:
is trained), from sample manually classified verbatims, the characteristics a
new verbatim should have in order to be attributed a given code.
The human operator who feeds sample manually classified verbatims to the
system plays the role of the “supervisor”.
Fabrizio Sebastiani (ISTI-CNR, Pisa, IT) ML & ATC: What’s Next? ASC Methods Conference 3 / 41
4. A Verbatim Coding System based on ML
Fabrizio Sebastiani (ISTI-CNR, Pisa, IT) ML & ATC: What’s Next? ASC Methods Conference 4 / 41
5. A Verbatim Coding System based on ML
Fabrizio Sebastiani (ISTI-CNR, Pisa, IT) ML & ATC: What’s Next? ASC Methods Conference 5 / 41
6. A Verbatim Coding System based on ML
Fabrizio Sebastiani (ISTI-CNR, Pisa, IT) ML & ATC: What’s Next? ASC Methods Conference 6 / 41
7. A Verbatim Coding System based on ML
Fabrizio Sebastiani (ISTI-CNR, Pisa, IT) ML & ATC: What’s Next? ASC Methods Conference 7 / 41
8. What I’ll be talking about today
A talk about the role of humans in the verbatim coding process, and about
how to best support their work
I will be looking at scenarios in which
1 automated verbatim coding technology is used ...
2 ... but the level of accuracy that can be obtained from the classifier is not
considered sufficient ...
3 ... with the consequence that one or more human coders are asked to inspect
(and correct where appropriate) a portion of the classification decisions, with
the goal of increasing overall accuracy.
Problem
How can we support / optimize the work of the human coders?
Fabrizio Sebastiani (ISTI-CNR, Pisa, IT) ML & ATC: What’s Next? ASC Methods Conference 8 / 41
9. What I’ll be talking about today
A talk about the role of humans in the verbatim coding process, and about
how to best support their work
I will be looking at scenarios in which
1 automated verbatim coding technology is used ...
2 ... but the level of accuracy that can be obtained from the classifier is not
considered sufficient ...
3 ... with the consequence that one or more human coders are asked to inspect
(and correct where appropriate) a portion of the classification decisions, with
the goal of increasing overall accuracy.
Problem
How can we support / optimize the work of the human coders?
Fabrizio Sebastiani (ISTI-CNR, Pisa, IT) ML & ATC: What’s Next? ASC Methods Conference 8 / 41
10. What I’ll be talking about today
A talk about the role of humans in the verbatim coding process, and about
how to best support their work
I will be looking at scenarios in which
1 automated verbatim coding technology is used ...
2 ... but the level of accuracy that can be obtained from the classifier is not
considered sufficient ...
3 ... with the consequence that one or more human coders are asked to inspect
(and correct where appropriate) a portion of the classification decisions, with
the goal of increasing overall accuracy.
Problem
How can we support / optimize the work of the human coders?
Fabrizio Sebastiani (ISTI-CNR, Pisa, IT) ML & ATC: What’s Next? ASC Methods Conference 8 / 41
11. What I’ll be talking about today
A talk about the role of humans in the verbatim coding process, and about
how to best support their work
I will be looking at scenarios in which
1 automated verbatim coding technology is used ...
2 ... but the level of accuracy that can be obtained from the classifier is not
considered sufficient ...
3 ... with the consequence that one or more human coders are asked to inspect
(and correct where appropriate) a portion of the classification decisions, with
the goal of increasing overall accuracy.
Problem
How can we support / optimize the work of the human coders?
Fabrizio Sebastiani (ISTI-CNR, Pisa, IT) ML & ATC: What’s Next? ASC Methods Conference 8 / 41
12. A worked out example
predicted
Y N
true
Y TP = 4 FP = 3
N FN = 4 TN = 9
F1 =
2TP
2TP + FP + FN
= 0.53
Fabrizio Sebastiani (ISTI-CNR, Pisa, IT) ML & ATC: What’s Next? ASC Methods Conference 9 / 41
13. A worked out example (cont’d)
predicted
Y N
true
Y TP = 4 FP = 3
N FN = 4 TN = 9
F1 =
2TP
2TP + FP + FN
= 0.53
Fabrizio Sebastiani (ISTI-CNR, Pisa, IT) ML & ATC: What’s Next? ASC Methods Conference 10 / 41
14. A worked out example (cont’d)
predicted
Y N
true
Y TP = 5 FP = 3
N FN = 3 TN = 9
F1 =
2TP
2TP + FP + FN
= 0.63
Fabrizio Sebastiani (ISTI-CNR, Pisa, IT) ML & ATC: What’s Next? ASC Methods Conference 11 / 41
15. A worked out example (cont’d)
predicted
Y N
true
Y TP = 5 FP = 2
N FN = 3 TN = 10
F1 =
2TP
2TP + FP + FN
= 0.67
Fabrizio Sebastiani (ISTI-CNR, Pisa, IT) ML & ATC: What’s Next? ASC Methods Conference 12 / 41
16. A worked out example (cont’d)
predicted
Y N
true
Y TP = 6 FP = 2
N FN = 2 TN = 10
F1 =
2TP
2TP + FP + FN
= 0.75
Fabrizio Sebastiani (ISTI-CNR, Pisa, IT) ML & ATC: What’s Next? ASC Methods Conference 13 / 41
17. A worked out example (cont’d)
predicted
Y N
true
Y TP = 6 FP = 1
N FN = 2 TN = 11
F1 =
2TP
2TP + FP + FN
= 0.80
Fabrizio Sebastiani (ISTI-CNR, Pisa, IT) ML & ATC: What’s Next? ASC Methods Conference 14 / 41
18. What I’ll be talking about (cont’d)
We need methods that
given a desired level of accuracy, minimize the human coders’ effort necessary
to achieve it; alternatively,
given an available amount of human coders’ effort, maximize the accuracy
that can be obtained through it
This can be achieved by ranking the automatically classified verbatims in
such a way that, by starting the inspection from the top of the ranking, the
cost-effectiveness of the human coders’ work is maximized
We call the task of generating such a ranking Semi-Automatic Text
Classification (SATC)
Fabrizio Sebastiani (ISTI-CNR, Pisa, IT) ML & ATC: What’s Next? ASC Methods Conference 15 / 41
19. What I’ll be talking about (cont’d)
We need methods that
given a desired level of accuracy, minimize the human coders’ effort necessary
to achieve it; alternatively,
given an available amount of human coders’ effort, maximize the accuracy
that can be obtained through it
This can be achieved by ranking the automatically classified verbatims in
such a way that, by starting the inspection from the top of the ranking, the
cost-effectiveness of the human coders’ work is maximized
We call the task of generating such a ranking Semi-Automatic Text
Classification (SATC)
Fabrizio Sebastiani (ISTI-CNR, Pisa, IT) ML & ATC: What’s Next? ASC Methods Conference 15 / 41
20. What I’ll be talking about (cont’d)
Previous work has addressed SATC via techniques developed for active
learning
In both cases, the automatically classified verbatims are ranked with the goal
of having the human coder start inspecting/correcting from the top; however
in active learning the goal is providing new training examples
in SATC the goal is increasing the overall accuracy of the classified set
We claim that a ranking generated “à la active learning” is suboptimal for
SATC1
1G Berardi, A Esuli, F Sebastiani. A Utility-Theoretic Ranking Method for Semi-Automated Text
Classification. Proceedings of the 35th Annual International ACM SIGIR Conference on Research and
Development in Information Retrieval (SIGIR 2012), Portland, US, 2012.
Fabrizio Sebastiani (ISTI-CNR, Pisa, IT) ML & ATC: What’s Next? ASC Methods Conference 16 / 41
21. What I’ll be talking about (cont’d)
Previous work has addressed SATC via techniques developed for active
learning
In both cases, the automatically classified verbatims are ranked with the goal
of having the human coder start inspecting/correcting from the top; however
in active learning the goal is providing new training examples
in SATC the goal is increasing the overall accuracy of the classified set
We claim that a ranking generated “à la active learning” is suboptimal for
SATC1
1G Berardi, A Esuli, F Sebastiani. A Utility-Theoretic Ranking Method for Semi-Automated Text
Classification. Proceedings of the 35th Annual International ACM SIGIR Conference on Research and
Development in Information Retrieval (SIGIR 2012), Portland, US, 2012.
Fabrizio Sebastiani (ISTI-CNR, Pisa, IT) ML & ATC: What’s Next? ASC Methods Conference 16 / 41
22. A Verbatim Coding System based on ML
Fabrizio Sebastiani (ISTI-CNR, Pisa, IT) ML & ATC: What’s Next? ASC Methods Conference 17 / 41
23. A Verbatim Coding System based on ML
Fabrizio Sebastiani (ISTI-CNR, Pisa, IT) ML & ATC: What’s Next? ASC Methods Conference 18 / 41
24. Outline of this talk
1 We discuss how to measure “error reduction” (i.e., the increase in accuracy
deriving from the human coder’s inspection activity)
2 We discuss a method for maximizing the expected error reduction for a fixed
amount of annotation effort
3 We show some promising experimental results
Fabrizio Sebastiani (ISTI-CNR, Pisa, IT) ML & ATC: What’s Next? ASC Methods Conference 19 / 41
25. Outline
1 Error Reduction, and How to Measure it
2 Error Reduction, and How to Maximize it
3 Some Experimental Results
Fabrizio Sebastiani (ISTI-CNR, Pisa, IT) ML & ATC: What’s Next? ASC Methods Conference 20 / 41
26. Error Reduction, and how to measure it
Assume we have
1 Code c;
2 Classifier h for c;
3 Set of unlabeled verbatims D that we have automatically classified by means
of h, so that every verbatim in D is associated
with a binary decision (Yes or No)
with a confidence score (a positive real number)
4 Measure of accuracy A, ranging on [0,1]
Fabrizio Sebastiani (ISTI-CNR, Pisa, IT) ML & ATC: What’s Next? ASC Methods Conference 21 / 41
27. Error Reduction, and how to Measure it (cont’d)
We will assume that A is
F1 =
2 · TP
(2 · TP) + FP + FN
but any measure of accuracy based on a contingency table may be used
An amount of error, measured as E = (1 − A), is present in the automatically
classified set D
Human coders inspect-and-correct a portion of D with the goal of reducing
the error present in D
Fabrizio Sebastiani (ISTI-CNR, Pisa, IT) ML & ATC: What’s Next? ASC Methods Conference 22 / 41
28. Error Reduction, and how to Measure it (cont’d)
We will assume that A is
F1 =
2 · TP
(2 · TP) + FP + FN
but any measure of accuracy based on a contingency table may be used
An amount of error, measured as E = (1 − A), is present in the automatically
classified set D
Human coders inspect-and-correct a portion of D with the goal of reducing
the error present in D
Fabrizio Sebastiani (ISTI-CNR, Pisa, IT) ML & ATC: What’s Next? ASC Methods Conference 22 / 41
29. Error Reduction, and how to Measure it (cont’d)
We define error at rank n (noted as E(n)) as the error still present in D after
the coder has inspected the verbatims at the first n rank positions
E(0) is the initial error generated by the automated classifier
E(|D|) is 0
We define error reduction at rank n (noted as ER(n)) to be
ER(n) =
E(0) − E(n)
E(0)
the error reduction obtained by the human coder who inspects the verbatims
at the first n rank positions
ER(n) ∈ [0, 1]
ER(n) = 0 indicates no reduction
ER(n) = 1 indicates total elimination of error
Fabrizio Sebastiani (ISTI-CNR, Pisa, IT) ML & ATC: What’s Next? ASC Methods Conference 23 / 41
30. Error Reduction, and how to Measure it (cont’d)
We define error at rank n (noted as E(n)) as the error still present in D after
the coder has inspected the verbatims at the first n rank positions
E(0) is the initial error generated by the automated classifier
E(|D|) is 0
We define error reduction at rank n (noted as ER(n)) to be
ER(n) =
E(0) − E(n)
E(0)
the error reduction obtained by the human coder who inspects the verbatims
at the first n rank positions
ER(n) ∈ [0, 1]
ER(n) = 0 indicates no reduction
ER(n) = 1 indicates total elimination of error
Fabrizio Sebastiani (ISTI-CNR, Pisa, IT) ML & ATC: What’s Next? ASC Methods Conference 23 / 41
31. Error Reduction, and how to Measure it (cont’d)
Fabrizio Sebastiani (ISTI-CNR, Pisa, IT) ML & ATC: What’s Next? ASC Methods Conference 24 / 41
32. Outline
1 Error Reduction, and How to Measure it
2 Error Reduction, and How to Maximize it
3 Some Experimental Results
Fabrizio Sebastiani (ISTI-CNR, Pisa, IT) ML & ATC: What’s Next? ASC Methods Conference 25 / 41
33. Error Reduction, and how to Maximize it
Problem
How should we rank the verbatims in D so as to maximize the expected error
reduction?
Fabrizio Sebastiani (ISTI-CNR, Pisa, IT) ML & ATC: What’s Next? ASC Methods Conference 26 / 41
34. A worked out example
predicted
Y N
true
Y TP = 4 FP = 3
N FN = 4 TN = 9
F1 =
2TP
2TP + FP + FN
= 0.53
Fabrizio Sebastiani (ISTI-CNR, Pisa, IT) ML & ATC: What’s Next? ASC Methods Conference 27 / 41
35. A worked out example (cont’d)
predicted
Y N
true
Y TP = 4 FP = 3
N FN = 4 TN = 9
F1 =
2TP
2TP + FP + FN
= 0.53
Fabrizio Sebastiani (ISTI-CNR, Pisa, IT) ML & ATC: What’s Next? ASC Methods Conference 28 / 41
36. A worked out example (cont’d)
predicted
Y N
true
Y TP = 5 FP = 3
N FN = 3 TN = 9
F1 =
2TP
2TP + FP + FN
= 0.63
Fabrizio Sebastiani (ISTI-CNR, Pisa, IT) ML & ATC: What’s Next? ASC Methods Conference 29 / 41
37. A worked out example (cont’d)
predicted
Y N
true
Y TP = 5 FP = 2
N FN = 3 TN = 10
F1 =
2TP
2TP + FP + FN
= 0.67
Fabrizio Sebastiani (ISTI-CNR, Pisa, IT) ML & ATC: What’s Next? ASC Methods Conference 30 / 41
38. A worked out example (cont’d)
predicted
Y N
true
Y TP = 6 FP = 2
N FN = 2 TN = 10
F1 =
2TP
2TP + FP + FN
= 0.75
Fabrizio Sebastiani (ISTI-CNR, Pisa, IT) ML & ATC: What’s Next? ASC Methods Conference 31 / 41
39. A worked out example (cont’d)
predicted
Y N
true
Y TP = 6 FP = 1
N FN = 2 TN = 11
F1 =
2TP
2TP + FP + FN
= 0.80
Fabrizio Sebastiani (ISTI-CNR, Pisa, IT) ML & ATC: What’s Next? ASC Methods Conference 32 / 41
40. Error Reduction, and how to Maximize it
Intuition 1: Verbatims that have a higher probability of being misclassified
should be ranked higher
Intuition 2: Verbatims that, if corrected, bring about a higher gain (i.e., a
bigger increase in A) should be ranked higher
This means that verbatims that have a higher utility (= probability × gain)
should be ranked higher
A false positive and a false negative may have different impacts on A !
While in active learning only the probability of misclassification is relevant, in
SATC gains are also relevant
Fabrizio Sebastiani (ISTI-CNR, Pisa, IT) ML & ATC: What’s Next? ASC Methods Conference 33 / 41
41. Error Reduction, and how to Maximize it (cont’d)
Given a set Ω of mutually disjoint events, a utility function is defined as
U(Ω) =
ω∈Ω
P(ω)G(ω)
where
P(ω) is the probability of occurrence of event ω
G(ω) is the gain obtained if event ω occurs
We can thus estimate the utility, for the aims of increasing A, of manually
inspecting a verbatim d as
U(TP, TN, FP, FN) = P(FP) · G(FP) + P(FN) · G(FN)
provided we can estimate
If d is labelled with code c: P(FP) and G(FP)
If d is not labelled with code c: P(FN) and G(FN)
Fabrizio Sebastiani (ISTI-CNR, Pisa, IT) ML & ATC: What’s Next? ASC Methods Conference 34 / 41
42. Error Reduction, and how to Maximize it (cont’d)
Given a set Ω of mutually disjoint events, a utility function is defined as
U(Ω) =
ω∈Ω
P(ω)G(ω)
where
P(ω) is the probability of occurrence of event ω
G(ω) is the gain obtained if event ω occurs
We can thus estimate the utility, for the aims of increasing A, of manually
inspecting a verbatim d as
U(TP, TN, FP, FN) = P(FP) · G(FP) + P(FN) · G(FN)
provided we can estimate
If d is labelled with code c: P(FP) and G(FP)
If d is not labelled with code c: P(FN) and G(FN)
Fabrizio Sebastiani (ISTI-CNR, Pisa, IT) ML & ATC: What’s Next? ASC Methods Conference 34 / 41
43. Error Reduction, and how to Maximize it (cont’d)
Estimating P(FP) and P(FN) (the probability of misclassification) can be
done by converting the confidence score returned by the classifier into a
probability of correct classification
Tricky: requires probability “calibration” via a generalized sigmoid function to
be optimized via k-fold cross-validation
Gains G(FP) and G(FN) can be defined “differentially”; i.e.,
The gain obtained by correcting a FN is (AFN→TP
− A)
The gain obtained by correcting a FP is (AFP→TN
− A)
Gains need to be estimated by estimating the contingency table on the
training set via k-fold cross-validation
Key observation: in general, G(FP) = G(FN)
Fabrizio Sebastiani (ISTI-CNR, Pisa, IT) ML & ATC: What’s Next? ASC Methods Conference 35 / 41
44. Error Reduction, and how to Maximize it (cont’d)
Estimating P(FP) and P(FN) (the probability of misclassification) can be
done by converting the confidence score returned by the classifier into a
probability of correct classification
Tricky: requires probability “calibration” via a generalized sigmoid function to
be optimized via k-fold cross-validation
Gains G(FP) and G(FN) can be defined “differentially”; i.e.,
The gain obtained by correcting a FN is (AFN→TP
− A)
The gain obtained by correcting a FP is (AFP→TN
− A)
Gains need to be estimated by estimating the contingency table on the
training set via k-fold cross-validation
Key observation: in general, G(FP) = G(FN)
Fabrizio Sebastiani (ISTI-CNR, Pisa, IT) ML & ATC: What’s Next? ASC Methods Conference 35 / 41
45. Error Reduction, and how to Maximize it (cont’d)
Estimating P(FP) and P(FN) (the probability of misclassification) can be
done by converting the confidence score returned by the classifier into a
probability of correct classification
Tricky: requires probability “calibration” via a generalized sigmoid function to
be optimized via k-fold cross-validation
Gains G(FP) and G(FN) can be defined “differentially”; i.e.,
The gain obtained by correcting a FN is (AFN→TP
− A)
The gain obtained by correcting a FP is (AFP→TN
− A)
Gains need to be estimated by estimating the contingency table on the
training set via k-fold cross-validation
Key observation: in general, G(FP) = G(FN)
Fabrizio Sebastiani (ISTI-CNR, Pisa, IT) ML & ATC: What’s Next? ASC Methods Conference 35 / 41
46. Outline
1 Error Reduction, and How to Measure it
2 Error Reduction, and How to Maximize it
3 Some Experimental Results
Fabrizio Sebastiani (ISTI-CNR, Pisa, IT) ML & ATC: What’s Next? ASC Methods Conference 36 / 41
47. Some Experimental Results
Dataset:
# Codes # Training # Test
Reuters-21578 115 9603 3299
Baseline: ranking by probability of misclassification (“à la active learning”),
equivalent to applying our ranking method with G(FP) = G(FN) = 1
Fabrizio Sebastiani (ISTI-CNR, Pisa, IT) ML & ATC: What’s Next? ASC Methods Conference 37 / 41
49. A few side notes
This approach allows the human coder to know, at any stage of the
inspection process, what the estimated accuracy is at that stage; obtained by
Estimating accuracy at the beginning of the process, via k-fold cross validation
Updating after each correction is made
This approach lends itself to having more than one coder working in parallel
on the same inspection-and-correction task
Recent research I have not discussed today :
A “dynamic” SATC method in which gains are updated after each correction
is performed
“Microaveraging” and “Macroaveraging” -oriented methods
Fabrizio Sebastiani (ISTI-CNR, Pisa, IT) ML & ATC: What’s Next? ASC Methods Conference 39 / 41
50. A few side notes
This approach allows the human coder to know, at any stage of the
inspection process, what the estimated accuracy is at that stage; obtained by
Estimating accuracy at the beginning of the process, via k-fold cross validation
Updating after each correction is made
This approach lends itself to having more than one coder working in parallel
on the same inspection-and-correction task
Recent research I have not discussed today :
A “dynamic” SATC method in which gains are updated after each correction
is performed
“Microaveraging” and “Macroaveraging” -oriented methods
Fabrizio Sebastiani (ISTI-CNR, Pisa, IT) ML & ATC: What’s Next? ASC Methods Conference 39 / 41
51. Concluding Remarks
Take-away message: Semi-automatic text classification needs to be addressed
as a task in its own right
Active learning typically makes use of probabilities of misclassification but does
not make use of gains ⇒ ranking “à la active learning” is suboptimal for SATC
The use of utility theory means that the ranking algorithm is optimized for a
specific accuracy measure ⇒ Choose the accuracy measure the best mirrors
your applicative needs (e.g., Fβ with β > 1), and choose it well!
SATC is important, since in more and more application contexts the accuracy
obtainable via completely automatic text classification is not sufficient; more
and more frequently humans will need to enter the loop
Fabrizio Sebastiani (ISTI-CNR, Pisa, IT) ML & ATC: What’s Next? ASC Methods Conference 40 / 41