The Data Blog

Key Papers about Episode Mining

Posted on 2021-08-16 by Philippe Fournier-Viger

This post presents the key papers about episode mining. If you are not familiar with what is episode mining, it is a data mining task, which aims at finding patterns in a sequence of events or symbols . A short introduction to episode mining is posted on my blog. Finding patterns in a sequence can be useful to discover regularities that may provide insights about the data, and even support decision making.

Below, I list the most important episode mining papers as a table with my comments on each paper to allow to get a quick overview of the field.

Note: that list is based on my analysis of the papers and is thus subjective and I may have missed some papers that other researchers would have deemed important.

Author and date	Paper title	Algorithm(s)	Key idea
Mannila et al. (1995)	Discovering frequent episodes in sequences	WINEPI, MINEPI	– This paper proposed the problem of episode mining. It is thus a key paper in that field. – A first algorithm called WINEPI finds episodes by performing a breadth-first search and using a sliding window. It counts the support (occurrence frequency) of an episode as the number of windows where the episode appears. However, WINEPIhas the problem that an occurrence may be counted more than once. – To address this issue, a second algorithm called MINEPI find frequent episodes by only considering the minimal occurrences of each episode. – This paper also presents a basic algorithm to generate episode rules by combining pairs of episodes. This is done as post-processing after applying WINEPI or MINEPI.
Huang et a. (2008)	Efficient mining of frequent episodes from complex sequences	EMMA, MINEPI+	– It is observed that window-based algorithms for episode mining such as MINEPI and WINEPI sometimes produce some strange results. – To address this issue, a novel measure is proposed to count the support (number of occurrences) of an episode called the head support or head frequency. – Two algorithms are designed to find frequent episodes using the head frequency: EMMA (a depth-first search algorithm) and MINEPI+ (a modified version of MINEPI)
Fournier-Viger et al. (2019)	TKE: Mining Top-K Frequent Episodes	TKE	– This paper makes the observation that it is difficult for users to set the minsup parameter for frequent episode mining. As a result users often have to spend considerable time fine tuning the parameters and may find too many or too few episodes. – As a solution this paper redefine the task of episode mining as top-k frequent episode mining, where the user can directly choose the number of patterns to be discovered (which is more intuitive). – The TKE algorithm is defined to efficiently find the top-k most frequent episodes.
Ao et al. (2018)	Online frequent episode mining		– This paper extended the concept of episode mining
Zhou et al. (2010)	Mining closed episodes from event sequences efficiently	Clo_episode	– This paper designed an algorithm to find closed episodes. The goal is to reduce the number of episodes presented by only showing some episodes that are said to be closed. This gives sometimes a small set of episodes that summarizes the frequent episodes. – An algorithm is presented to find closed episodes called Clo_episode. This algorithm adopts a breadth-first search and use the concept of minimal occurrences. – A limitation: this paper cannot handle the case of multiple events happening at the same time (parallel episodes). while most algorithms can handle this case.
Tati and Cule (2011)	Mining closed episodes with simultaneous events		– This paper presents algorithms to find closed episodes where events may be simultaneous. – This fixes the main limitation of the paper of Zhou et al. (2010).
Laxman et al.(2007)	Discovering frequent episodes and learning Hidden Markov Models: A formal connection		– Laxman proposes to only count the non overlapping occurrences of episodes. – This counting method is called the non overlapping support or non overlapping frequency, and was then used in many other papers after.
Laxman et al. (2007)	A Fast Algorithm For Finding Frequent Episodes In Event Streams	Algorithm 1 and Algorithm 2	– This paper introduces an algorithms to find frequent episodes in a potentially infinite sequence of events (a data stream)
Oualid et al. (2021)	Mining Episode Rules from Event Sequences Under Non-overlapping Frequency	NONEPI	– This paper presents an algorithm named NONEPI for episode rule mining using the concept of non overlapping frequency. – The goal is to find rules that are easier to interpret as occurrences must be non overlapping.
Fournier-Viger et al (2021)	Mining Partially-Ordered Episode Rules in an Event Sequence	POERM	– This paper makes the observation that traditional episode rules have a very strict ordering between events. – This paper defines are more general type of episode rules called partially-ordered episode rules. These rules loosen the ordering constraint between events. As a result, a partially-ordered episode rule can summarize multiple traditional episode rules. – The POERM algoriths is defined to find partially-ordered episode rules. It finds rules directly without first having to apply a frequent episode mining algorithm. – Another version of POERM called POERM_H was proposed in a subsequent paper “Mining Partially-Ordered Episode Rules with the Head Support“, where occurrences are counted using the head support of EMMA.
Fahed et al. (2018)	DEER: Distant and Essential Episode Rules for early prediction	DEER	– This paper presented an algorithm named DEER to find episore rules that can predict distant events rather than events that appear closely after other events. These rules are called essential rules. – The algorithm is based on the concept of minimal occurrences. – Limitation: the algorithm does not handle the case of simultaneous events.
Wu et al. (2013)	Mining high utility episodes in complex event sequences	US-SPAN	– A problem with traditional episode mining algorithm is that all event types are considered as equally important. – To address this issue, the concept of utility is added to episode mining to define a new problem of high utility episode mining. – In that problem each event occurrence can have a quantity as well as a weight. This allows for example to model the purchases of customers with quantities and unit prices to find episodes that yield the most money (the highest utility). – An efficient algorithm called US-SPAN is proposed for this problem, which is based on the concept of minimal occurrences.
Fournier-Viger et al. (2019)	HUE-Span: Fast High Utility Episode Mining	HUE-SPAN	– This paper makes the important observation that the previous algorithm for high utility episode mining US-SPAN can underestimate the utilitiy of episodes by not taking into account all timestamps of minimal occurrences for utility calculations. As a result, some high utility episodes can be missed. – To address this issue, the definition of utility is modified. – Moreover, a new and more efficient algorithm named HUE-SPAN is proposed for high utility episode mining. – This algorithm is based on the concept of minimal occurrences.
Ao et al. (2018)	Large-scale Frequent Episode Mining from Complex Event Sequences with Hierarchies	LA-FEMH	– A big data algorithm for episode mining called LA-FEMH is proposed using the Spark architecture. – The algorithm can find closed and maximal episodes and also consider that events are organized as a taxonomy. – Limitation: This algorithm does not handle the case of simultaneous events. In other words, the algorithm can only find serial episodes.
Fournier-Viger et al. (2022)	MaxFEM: Mining Maximal Frequent Episodes in Complex Event Sequences	MaxFEM, AFEM	– An algorithm called MaxFEM to find maximal episodes in a complex event sequence (the general case). – A version called AFEM to find all frequent episodes. – Those extends the EMMA algorithm with new optimizations and use the head support definition.

Implementations

There are very few software programs and source code available online for episode mining. The most complete software, which offers a dozen episode mining algorithms and is open-source is the SPMF data mining software (which I am the founder). It provides implementations of many algorithms such as MINEPI, EMMA, TKE， US-SPAN, POERM, and HUE-SPAN.

Besides episode mining, SPMF also offers algorithms for many other data mining tasks such as sequence prediction, high utility itemset mining, sequential pattern mining, periodic pattern mining, sequential rule mining and subgraph mining.

Survey paper on episode mining

If you want to read a detailed survey paper about episode mining, you can also check this survey paper:

Ouarem, O., Nouioua, F., Fournier-Viger, P. (2023). A Survey of Episode Mining. WIREs Data Mining and Knowledge Discovery, Wiley, to appear.

Conclusion

In this blog post, I have given a list of key papers about episode mining. Of course, making such list is subjective. But I believe that this list can be useful for those who wants to learn quickly about episode mining by having a quick summary.

If you have enjoyed this blog post, you may also check other content of this blog. There are many posts and resources related to pattern mining.

Philippe Fournier-Viger is a full professor working in China and founder of the SPMF open source data mining software.

What is Machine Learning?

Posted on 2021-08-09 by Philippe Fournier-Viger

In recent years, machine learning has become a popular research area of computer science. In this blog post, I will talk about what is machine learning.

What is machine learning?

Machine learning is computer programs (algorithms) that can learn to do a task by doing it or by learning it from some data. For example, a computer program can be designed to learn to play Chess by trying various moves and strategies against human players to select the most effective one (learning by experience). Or a program could be designed to learn the best Chess tactics from historical records of matches between human players (data).

What is the difference with artificial intelligence?

Another term that is often talked about in the media is artificial intelligence. So what is the difference between artificial intelligence and machine learning? Generally, artificial intelligence refers to computer programs that can do some task that requires intelligence (e.g. translating a document from English to French, writing a summary of a text, playing a game, composing some music). Some artificial intelligence program will be designed to learn from experience or data, and can thus be viewed as using machine learning. But there are also some other artificial intelligence programs that do not require learning. For example, one can build a program to play the game of Tic Tac Toe by explicitly writing the optimal rules for playing that game in the program. In this case, the program can use that knowledge base (rules) to play and does not need to learn. Thus, not all artificial intelligence programs need to learn, and machine learning can be viewed as a subset of artificial intelligence that represents the programs learn.

Why is machine learning popular?

There exists many machine learning techniques for making computer programs that can learn such as artificial neural networks, support vector machines and clustering. These techniques are popular because they can be used to create programs that learn to do some complicated tasks that would be otherwise very hard to build as a computer program by a human. For example, while it is extremely hard to program a computer to recognize objects in a video by hand or to translate a text accurately, these tasks can now be learned using machine learning techniques.

What are the applications of machine learning?

There are many applications of machine learning such as to play games, process video and audio data, translate documents, and recommend songs or movies to users of a website. Generally, each machine learning program is built to solve a specific task (e.g. playing the game of Go) rather than to solve many tasks. It is a major challenge to design machine learning programs that could learn many tasks.

What are some good books about machine learning?

Nowadays, many young researchers will directly focus on popular techniques such as deep learning. But in my opinion, one should try to have a broader picture of the field of machine learning as there are many other techniques. Some good book on machine learning in general are:

Pattern Recognition and Machine Learning by C. Bishop
Machine Learning by T. Mitchell
The Elements of Statistical Learning: Data Mining, Inference, and Prediction, Hastie et al.
Machine Learning: A Probabilistic Perspective – K. Murphy
Artificial Intelligence: A Modern Approach – Russell & Norvig

If you are a young researcher, you may also want to read about how to find a good machine learning research topic.

What is the relationship with data mining, big data and data science?

Another subfield of computer science that is popular nowadays is data mining, data science and big data. These terms generally refer to the use of algorithms to analyze data. There are generally two main goals to analyze the data: (1) understanding the data to learn something useful from the data (e.g. understanding the past, like why a tsunami occurred) and (2) predicting the future (e.g. predicting when the next tsunami will hit a country).

The data mining techniques that aim at making predictions using data can be viewed as a form of machine learning, while other techniques may just be viewed as method to analyze data. Thus, data mining can be viewed as intersecting with machine learning. If you want to know more about this, I wrote a blog post on the relationship between data mining and machine learning.

Conclusion

Hope that this blog post has been interesting! If you have any comments, please leave them in the comment section below.

—-

Philippe Fournier-Viger is a full professor working in China and founder of the SPMF open source data mining software.

Posted in artificial intelligence, Machine Learning | Tagged artificial intelligence, data mining, machine learning, ml | 8 Comments

Why Journal Special Issues are Popular?

Posted on 2021-08-02 by Philippe Fournier-Viger

In this blog post, I will talk about journal special issues. I will talk about how special issues of journals are organized and why it can be good to submit to or organize academic journal special issues.

What is a special issue of a journal?

First, I will explain some basic concepts about journals. An academic journal publishes papers written by researchers. These papers are generally published as part of an issue and a volume. An issue is a collection of multiple papers that may not be on the same topic, while a volume is a collection of one or more issues.

For example, a journal may publish an issue every month and four volumes per year. Thus, the first volume would contains the issues of January, February, March and April. Then, the second volume would contain the issues of May, June, July and August, and so on.

Volumes and issues are generally numbered as 1,2,3…. But it is to be noted that some journals do not follow this. For example, some journal will name volumes according to years such as volume 2021 for the year 2021. Moreover, some journals only group papers by volumes and don’t have issue numbers.

Having explained that, what is a special issue? A special issue is a group of papers that are generally on the same topic. For example, a journal may publish a special issue on pattern mining where all the papers are about this topic. There are also some conferences that will organize special issues in a journal that will contain the best papers of the conference. In that case, the paper of the special issue may not be on the same topic but are grouped based on another criterion.

Each journal has one or more editors that take care of managing the review process of papers by doing some tasks such as inviting reviewers, reading the recommendations of reviewers and deciding to accept, revise or reject papers. For special issues, the papers are generally handled by some guest editors rather than the regular editor(s) of the journal. This means that some researchers are responsible of the special issue, and those researchers are called “guest” because they do not usually work for the journal.

How are the special issues organized?

Generally, the special issues are organized by researchers that talk with the main editor of the journal and ask to organize a special issue in that journal on a specific topic. Then, if the editor accepts, the special issue will be created and there will be some advertisement to invite authors to submit papers to the special issue. A special issue may be open for several months, which means that authors may have several months to write a paper and submit it to a given special issue. When a paper is received for a special issue, it is the guest editors of the special issue that will organize its review.

To propose a special issue, several journals will require to submit a special issue proposal explaining the topic of the special issue, the reasons why the topic is timely, the background of the guest editors, etc. Some journals will receive many such proposals and will only organize a few special issues.

Why organizing special issues?

Journals typically organize special issues to attract papers on emerging or trending topics, as this is what will bring more papers and also citations. For example, a journal may organize a special issue about machine learning for analyzing the COVID19 genome as COVID is now a popular topic. Generally, journals do not like to organize special issues on old topics.

For a journal, special issues can help to bring more papers. This is the reason why some young journals will organize many special issues, while some top journals will rarely organize special issues. Thus a young journal is more likely to accept a proposal for making a special issue than a popular journal, who may just ignore them.

Why researchers want to organize special issues?

There are several reasons. First, it shows that a researcher is able to organize things and it gives editorial experience to a researcher about how to manage papers in a journal. This may latter help a researcher when applying to organize other things such as a workshop, a book or a position in a journal. Second, it gives visibility to a researcher and may help build connections with other researchers.

There are also some researchers that will try to abuse the concept of special issue by organizing many special issues and accepting the papers of their friends in exchange for their papers being in turn accepted in other special issues. This is something unethical that some people do. I have noticed this on websites like DBLP. I will not give any names but I found that some researchers always publish in some special issues of their friends. This does not look good and should be avoided. A researcher should always avoid conflicts of interest and be honest.

Why researchers publish in special issues?

A main reason for a researcher to publish in a special issue is to be part of an issue where all papers are on the same main topic. This will thus give visibility to the paper. For example, a special issue on periodic pattern mining may publish numerous papers on that topic.

Another reason why a researcher may choose to publish in a special issue is that often the review process is faster and sometimes there is a higher chance for a paper to be accepted than if the paper was submitted as a regular paper to the journal. This is for example often the case for special issues about the best papers of a conference. In that case, maybe 14 papers may be invited and 10 or more will be accepted.

Conclusion

In this blog post, I talked about the concept of special issue for academic journals. I gave an overview about why publish in and organize special issues can be interesting for researchers. Hope it has been interesting. Feel free to leave some comments below!

—

Philippe Fournier-Viger is a full professor working in China and founder of the SPMF open source data mining software.

Posted in Academia, General, Research | Tagged academia, editor, guest editor, journal, Research, special issue | Leave a comment

Useful Latex tricks for Writing Research Papers

Posted on 2021-07-27 by Philippe Fournier-Viger

In this blog post, I will talk about some useful latex tricks for researchers writing research papers using Latex. This blog post is aimed at those who knows already how to use Latex but maybe do not know these tricks.

1.Reducing the length of your paper with \vspace

A common problem when writing a research paper is that the paper is too long. Besides rewriting the text to make it shorter, a solution is to use some special Latex commands to reduce the space. WARNING: But be aware that it is sometimes forbidden to use these commands, so use them at your own risk!

The main command is \vspace. It allows to reduce the vertical space between elements on a page. For example using \vspace{-0.5cm} before a figure will reduce the space before that figure of 0.5 cm. This is a very useful command. But it is recommended to use it after finishing writing a paper as this command can easily mess up the layout of your paper if the content is then changed.

2. Reducing the length of an algorithm written using algorithm2e

Another way of reducing the space in a paper is to reduce the size of an algorithm. A command that can be used is \scriptsize after \begin{algorithm}. This will reduce the font size of the algorithm and thus the space.

If you are using the algorithm2e package for your algorithms, another way of reducing the length of an algorithm is to use an inline IF instead of a regular IF. This is done by replacing \if{} by \lIf{}. The result is:

This can save a few lines. Similarly, it is possible to replace a \forEach{} loop by the inline version \lForEach{}. Oher algorithm2e commands can also be used as inline such as \else and \lElse.

Another useful command to reduce the size of an algorithm written with algorithm2e is to use \SetAlgoNoEnd after \begin{algorithm}. This will remove the “END” labels for all the IF, ELSE and FOR EACH parts. For example, the below picture show the effect:

3. Check if your paper contains uncited references with \refcheck

If you want to quickly find all the references that are not cited in your paper, you just need to add this: \usepackage{refcheck}. It will higlight the references that are not used from your bibliography. For example:

4. Comparing two versions of your LaTeX document with Latexdiff

Another very useful tool is LatexDiff. Many journals will ask authors to highlight the differences between two versions of their papers. I previously wrote a detailled blog post about using LatexDiff. Please see that blog post for details. The result is like this:

5. Adding TODO notes

Another useful tool is the TODONOTES package. It allows to add TODO comments on a latex document. This works well with the IEEE template. For example, by adding \usepackage{todonotes}, we can add comments in the document such as \todo{Error!} and it will appear like this:

6. Adding color to your Latex document

Another useful package is the color package. It allows to change the color of some part of your document. This can be useful to highlight what remains to be done in your paper or what has should be revised.

7. Converting Latex to HTML

Sometimes, you may want to convert your Latex paper to an HTML document. You may have a look at my previous blog post on this topic to see how to do it with HTLATEX.

Conclusion

In this blog post, I wanted to share a few useful Latex commands. If you think I have missed some other important commands (surely!), please share in the comment section below. I might then add them to the blog post.

—

Philippe Fournier-Viger is a full professor working in China and founder of the SPMF open source data mining software.

Posted in Latex, Research | Tagged latex, Research, research paper, writing | Leave a comment

Brief report about the IEA AIE 2021 conference

Posted on 2021-07-26 by Philippe Fournier-Viger

This week, it is the IEA AIE 2021 conference (34th Intern. Conf. on Industrial, Engineering & Other Applications of Applied Intelligent Systems), which is held from 26th to 28th June 2021. This year, the conference is held online due to the COVID pandemic situation around the world.

In this blog post, I will give an overview of the conference.

About IEA AIE 2021

The IEA AIE conference is a conference that focuses on artificial intelligence and its applications. I have attended this conference several times over the year. I have written some blog posts also about IEA AIE 2016, IEA AIE 2018, IEA AIE 2019 and IEA AIE 2020.

This year, there has been 145 papers submitted. From this, 87 papers were accepted as full papers, and 19 as short papers.

Special sessions

This year, there was eight special sessions organized at IEA AIE on some emerging topics. A special session is a special track for submitting papers, organized by some guest researchers. All accepted papers from special sessions are published in the same proceedings as regular papers.

Special Session on Data Stream Mining: Algorithms and Applications
(DSMAA2021)
Special Session on Intelligent Knowledge Engineering in Decision Making Systems
(IKEDS2021)
Special Session on Knowledge Graphs in Digitalization Era (KGDE2021)
Special Session on Spatiotemporal Big Data Analytics (SBDA2021)
Special Session on Big Data and Intelligence Fusion Analytics (BDIFA2021)
Special Session on AI in Healthcare (AIH2021)
Special Session on Intelligent Systems and e-Applications (iSeA2021)
Special Session on Collective Intelligence in Social Media (CISM2021).

Opening ceremony

On the first day, there was the opening ceremony. It was announced that IEA AIE 2022 will be held in Japan next year.

Keynote speakers

There was two keynote speakers: (1) Prof. Vincent Tseng from National Yang Ming Chiao Tung University, (2) Prof. Francisco Herrera from University of Granada.

Paper presentations

I have attended several paper presentations through the conference. There was some high quality papers on various topics related to artificial intelligence. There was four rooms with paper presentations. Here is a screenshot of one of the rooms:

In particular, this year, there was six papers on pattern mining topics such as high utility pattern mining, sequential pattern mining and periodic pattern mining:

Oualid Ouarem, Farid Nouioua, Philippe Fournier-Viger: Mining Episode Rules from Event Sequences Under Non-overlapping Frequency. 73-85
Comment: This paper presents a novel algorithm for episode rule mining called NONEPI. The idea is to find rules using the non-overlapping frequency in a sequence of events.
Sumalatha Saleti, Jaya Lakshmi Tangirala, Thirumalaisamy Ragunathan: Distributed Mining of High Utility Time Interval Sequential Patterns with Multiple Minimum Utility Thresholds. 86-97
Comment: This paper presents a new algorithm DHUTISP-MMU for mining high utility time interval sequential patterns with multiple minimum utility thresholds. A key idea in this paper is to add information about the time intervals between items of sequential patterns. Besides, the algorithm is distributed.
Xiangyu Liu, Xinzheng Niu, Jieliang Kuang, Shenghan Yang, Pengpeng Liu: Fast Mining of Top-k Frequent Balanced Association Rules. 3-14
Comment: This paper presents an algorithm named TFBRM for mining the top-k balanced association rules. There has been a few algorithms for top-k association rule mining in the bast. But here a novelty is to combine support, kulczynski (kulc) and imbalance ratio (IR) as measures to find balanced rules.
Penugonda Ravikumar, Likhitha Palla, Rage Uday Kiran, Yutaka Watanobe, Koji Zettsu: Towards Efficient Discovery of Periodic-Frequent Patterns in Columnar Temporal Databases. 28-40
Comment: This paper presents an Eclat-based algorithm for periodic pattern mining called PF-Eclat. From the presentation it seems to me that this algorithm is very similar to the PFPM algorithm (2016) that I proposed 5 years ago. The difference seems to be that the vertical representation is a list of timestamps instead of list of TIDs, and it has two less constraints. That is the user can only use maxPer and minSup(minAvg) as constraints but PFPM also offers two more constraints: minPer and maxAvg. By the way, there exists also another Eclat based algorithm for a similar task (mining top-k periodic frequent patterns) called MTKPP (2009).
Sai Chithra Bommisetty, Penugonda Ravikumar, Rage Uday Kiran, Minh-Son Dao, Koji Zettsu: Discovering Spatial High Utility Itemsets in High-Dimensional Spatiotemporal Databases. 53-65
Tzung-Pei Hong, Meng-Ping Ku, Hsiu-Wei Chiu, Wei-Ming Huang, Shu-Min Li, Jerry Chun-Wei Lin: A Single-Stage Tree-Structure-Based Approach to Determine Fuzzy Average-Utility Itemsets. 66-72
Comment: This paper is about fuzzy high utility itemset mining. A novel algorithm is presented. A difference also with previous paper is the use of the average utility function in fuzzy high utility itemset mining.

Next year

The IEA AIE 2022 conference will be held in Kitakyushu, Japan.

Conclusion

This was a good conference. I have attended several presentations and had a chance to discuss with some interesting researchers. Looking forward to the IEA AIE 2022 conference.

Philippe Fournier-Viger is a full professor working in China and founder of the SPMF open source data mining software.

Posted in artificial intelligence, Conference, Machine Learning | Tagged ai, artificial intelligence, conference, iea aie, ieaaie, ieaaie2021, machinelearning | 7 Comments

Brief report about the DSIT 2021 conference (4th Intern. Conf. on Data Science and Information Technology)

Posted on 2021-07-23 by Philippe Fournier-Viger

This week, I am attending the DSIT 2021 conference (4th International Conference on Data Science and Information Technology) from July 23 to 25 in Shanghai, China.

The DSIT 2021 conference is co-located with the DMBD 2021 conference (the 4th International Conference on Data Mining and Big Data).

DSIT is a relatively young conference, which focuses on data science and data mining. But the quality was good and it was well organized. The proceedings of the conference are published by ACM. Thus, all papers are in the ACM Digital Library. This gives visibility to the papers.

A total of 150 submissions were received and 80 full papers were accepted for publication (acceptance rate = 53%). The papers were from several countries including China, Japan, Singapore, Vietnam, Philippines, Pakistan, Thailand, USA, Greece, France and Germany.

There was also several keynote speakers: Prof. Tok Wang Ling from National University of Singapore, Prof. Ma Maode from Nanyang Techn. University of Singapore, Prof. Shigeo Akashi from Tokyo University of Science, Japan and Prof. Philippe Fournier-Viger (myself) from Harbin Inst. of Technology (Shenzhen), China.

Due to the COVID pandemic and travel restrictions, the conference was held in Shanghai but some speakers were online through Zoom.

Day 1 – Registration

On the first day, I registered at the conference reception desk at hotel and receive a bag with the program, ID card, a small gift, and other things.

Day 2 – Keynote Talk

First, there was the opening ceremony.

Then, it was the keynote talks. I started first with my invited talk on algorithms for discovering patterns in data that are in interpretable (pattern mining).

Then, there was the talk by Prof. Jie Yang on adversarial attacks on deep neural networks. He has shown some recent work on generating adversarial pictures to fool neural networks. For instance a picture of a car may be slightly modified to fool a neural network into believing it is a house. What I find the most interesting about this talk is that it was shown that some modified pictures can fool not only one network but all the state of the art deep neural networks for image recognition. The reason why it is possible to fool multiple networks with a same modified picture is that an attack based on attention was used and that many deep neural networks will use attention in a similar way (focusing on the same image features). A dataset of adversarial images called DAmageNet was also presented, which can be helpful to test ways to protecting against such attacks. An interesting conclusion was that these attacks are possible because deep neural models tend to ignore some important features and incorporate unnecessary features.

Then, there was the other keynote talks.

Day 2 – Paper presentation

Then there was the regular paper presentations and a poster session.

There was two papers related to pattern mining. The first one was about high utility itemset mining and the other about frequent pattern mining.

High Utility pattern mining based on historical data table over data streams by Xinru Chen, Pengjun Zhai and Yu Fang
MaxRI: A method for discovering maximal rare itemsets by Sadeq Darrab et al.

I took some pictures of a few slides from that paper about maximal rare itemsets, as I find this to be an interesting topic:

Conclusion

This is all I will write for this conference. Overall, that was an interesting conference. It is not a very big conference but I met some other interesting researchers and we had some good discussions. Some papers were also quite good.

In a few days, I will be attending the IEA AIE 2021 conference and will report also about it.

—

Philippe Fournier-Viger is a full professor working in China and founder of the SPMF open source data mining software.

Posted in artificial intelligence, Big data, Conference, Data Mining | Tagged artificial intelligence, china, data, data mining, data science, dmbd, dsit, machine learning | 2 Comments

Brief report about the CCF-AI 2021 conference

Posted on 2021-07-21 by Philippe Fournier-Viger

This week, I attend the CCF-AI 2021 conference, which is the Chinese Computer Federation conference on Artificial Intelligence. This conference is held in the city of Yantai (烟台) in Shandong province of China, from the 22th to 24th July 2021.

About CCF-AI

CCF-AI is a national conference. But it is a major conference in China, with over 1,000 attendees. I attend this conference to meet other researchers and get to know about the recent results in this area. There are many high level speakers at the conference and activities.

In the past CCF-AI has been held in various locations around China. Here is a few of them:

Location

The city where CCF-AI is held this year is Yantai (烟台). It is a coastal city in eastern China, in Shandong province. It has good weather during the summer, beaches and many other activities.

The conference was held at the Yantai International Expo Center:

Registration

After arriving at the hotel, all attendees have to pass a test for the COVID to ensure the safety of everyone at the conference. Then, I registered and received my bag and badge with the program and other information.

Day 1 – Multi-Agent Systems forum

The conference is divided into some sub-forums. On the morning of the first day, I attended the multi-agent system forum. I also had some good discussions with other researchers.

Day 1 – Meeting of CCF-AI members

On the evening, I attended the meeting of CCF-AI members.

It was voted that CCF-AI 2023 will be held at Xinjiang University in Urumqi, China.

There was also a vote to select new members of CCF-AI. I am happy to have been selected:

It was said that for CCF–AI 2021, 339 papers were submitted and 128 papers were accepted (38% acceptance rate).

Other days and conclusion

There was also many other interesting activities and talks at this conference in the following days. However, my schedule was very tight. I came to CCF-AI, right after attending ICSI 2021, and I had to leave on the second day of CCF-AI to go to Shanghai to attend the DSIT 2021 conference in Shanghai, which I will talk about in the next blog post! Then, I will also attend the IEA AIE 2021 conference.

—

Philippe Fournier-Viger is a full professor working in China and founder of the SPMF open source data mining software.

Posted in artificial intelligence, Conference, Machine Learning | Tagged 2021, ai, artificial intelligence, ccf, ccfai, ccfai 2021, china, conference, machine learning, ml, yantai | 1 Comment

Brief report about ICSI 2021 (12th Int. Conference on Swarm Intelligence)

Posted on 2021-07-15 by Philippe Fournier-Viger

In this blog post, I will talk about attending the 12th International Conference on Swarm Intelligence (ICSI 2021). The ICSI conference is a relatively young conference about swarm intelligence, metaheuristics and related topics and applications. This year, ICSI 2021 is held in Qingdao, a coastal city in eastern China, from July 17–21, 2021. The conference is also held partially online for those that cannot attend due to travel restrictions.

The conference was held at the Blue Horizon Hotel:

The ICSI conference has been held in several cities and countries, over the years:

ICSI 2020 – Serbia (virtual)
ICSI 2019 – Chiang Mai, Vietnam
ICSI 2018 – Shanghai, China
ICSI 2017 – Fukuoka, Japan
ICSI 2016 – Bali, Indonesia
ICSI-CCI 2015 – Beijing, China
ICSI 2014 – Hefei, China
ICSI 2013 – Harbin, China
ICSI 2012 – Shenzhen, China
ICSI 2011 – Chongqing, China
ICSI 2010 – Beijing, China

Proceedings

The proceedings of the ICSI conference are published in the Springer Lecture Notes in Computer Science (LNCS) series as two volumes (Part 1 and Part 2). This ensures that the proceedings are indexed by EI and other indexes like DBLP.

This year, the conference received 177 submissions, which were reviewed on average by 2.5 reviewers. From this 104 papers were accepted for publications, which means an acceptance rate of 58.76%. The paper were organized into 16 sessions.

Day 1 – Registration

On the first day, I registered. I received a paper bag with a badge and the conference program. The proceedings was available online as a download.

Day 1 – Reception

There was also a reception at the hotel in the evening that lasted about an hour. There was food, beer and other drinks. This was a social activity, which is a good opportunity to discuss with other researchers that attend the conference.

Day – 2 – Opening ceremony
On the second day there was the opening ceremony, where the general chair talked about the conference, and the program.

The program committee chair also talked about the paper selection process.

Day – 2 – Keynote talks and invited talks

On the second day, there was two keynote talks and two invited talks. Some good researchers had been invited, and some of the talks were quite interesting. Below is a very brief overview.

The first keynote talk was by Prof. Qirong Tang from Tongji University who talked about “Large-Scale Heterogeneous Robotic Swarms”. He developed a swarm robotic platform that is used for some applications such as searching for multiple light sources, searching for a target, drug delivery in the body, etc. The idea is that some robots can cooperate together to perform a task more quickly (e.g. cooperative search) and thus outperform a single high quality robot. The swarm can be heterogeneous, that is using different types of robots such as flying robots and ground robots. Many bio-inspired algorithms are used to control a robot swarm such as particle swarm optimization (PSO) and genetic algorithms but it was argued that PSO is particularly suited for this task.

The second keynote talk was online by Prof. Chaomin Luo from USA about swarm intelligence applications to robotics and autonomous systems. This includes for example, exploration robots, search and rescue robots.

There was an invited talk by Prof. Gai-Ge Wang from Ocean university. He talked about how to improve the performance of metaheuristics using information feedback. The idea is that during iterations, some feedback of previous iterations is used to guide the search process towards better solutions.

The second invited talk was by Prof. Wenjian Luo from Harbin Institute of Technology (Shenzhen) about many-objectives optimization when multiple parties are involved. For example, to buy a car, many objectives may have to be considered such as the price, size, and fuel consumption and multiple parties such as an husband and wife may put different weights on those objectives. The goal is to find a solution that is optimal for all the parties involved but it is not always possible.

Day 2 – Paper presentations

On the afternoon, there was paper presentations and a poster session. There was some good papers about a variety of topics such as sheep optimization, classification of imbalanced data with PSO, citation analysis, swarm intelligence for UAVs, and multi-robot cooperation.

I have presented the below data mining paper about proof searching for proving theorems using simulated anneealing (which is mainly the work of my post-doc. M. S. Nawaz). In that paper, we use the simulated annealing metaheuristic to search for proofs to PVS theorems and compare with a genetic algorithm.

Nawaz, M. S., Sun, M., Fournier-Viger, P. (2021). Proof Searching in PVS theorem prover using Simulated Annealing. Proceedings of the 12th Intern. Conf. on Swarm Intelligence (ICSI 2021) Part II, pp. 253-262

There was also a good paper by Prof. Wei Song et al. about using fish swarm optimization for high utility itemset mining:

Song, W. Li, J. Huang, C.: Artificial Fish Swarm Algorithm for Mining High Utility Itemsets. ICSI (2) 2021: 407-419

Day 2 – Banquet

In the evening, there was a banquet. The best paper awards were announced.

ICSI 2022
It was announced that next year the ICSI 2022 conference will be held in Xian, China from July 15 to 19 2022.

Conclusion
Swarm intelligence is not my main research area although I have participated to several papers on this topic. But the conference was interesting and well organized. The quality was generally good. I would attend it again if I have some papers on this topic.

Now, I will leave Qingdao, and next I will attend the CCF-AI 2021 conference, DSIT 2021 conference, and then the IEA AIE 2021 conference.

—
Philippe Fournier-Viger is a professor of Computer Science and also the founder of the open-source data mining software SPMF, offering more than 120 data mining algorithms.

Posted in artificial intelligence, Conference, Data Mining, Data science | Tagged ai, artificial intelligence, china, icsi, icsi conference, machine learning, metaheuristics, optimization, Qingdao, swarm intelligence | 2 Comments

SPMF 2.48

Posted on 2021-06-28 by Philippe Fournier-Viger

Hi all, I have not been very active on the blog during the last month. This is because I had many thinsg going on in my personal and professional life that I will not reveal here. But I will be back soon with more regular content for the blog. Today, I write a blog post to give you some news:

SPMF 2.48

First, I would like to say that a new version of SPMF data mining software has just been released (v. 2.48) with two new algorithms:
– NEclatClosed for mining closed itemsets
– HUIM-SPSO for mining high utility itemsets using Set-based Particle Swarm Optimization
Those are the original implementations, provided by the authors.

T com.

MLiSE 2021 – deadline extension

Third, I would like to mention that the deadline for submiting your papers to the MLiSE 2021 workshop at PKDD that I co-organize has been extended to the 15th July. The theme of the workshop is Machine Learning in Software Engineering but the scope can be more broad so if you have any question about the workshop, feel free to contact with me. I would be happy to see your paper 🙂

Conclusion

This blog post was just to give some quick update. Hope it has been interesting.

—
Philippe Fournier-Viger is a professor of Computer Science and also the founder of the open-source data mining software SPMF, offering more than 200 data mining algorithms.

Posted in Machine Learning, Pattern Mining, spmf, Website | Tagged mlise, pattern mining, pattern mining wiki, spmf, wiki | Leave a comment

Brief report about ICIVIS 2021 (Int. Conference on Image, Vision and Intelligent system)

Posted on 2021-06-19 by Philippe Fournier-Viger

This week-end, I have attended the International Conference on Image, Vision and Intelligent system from 18 to 20 June 2021 in Changsha city, China.

It is a medium-sized conference (about 100 participants) but It is well-organized, and there was many interesting activites and speakers, as well as some workshops. The main theme of this conference is about image and computer vision but also some other works more related to intelligent systems where presented.

I have participated to this conference as an invited keynote speaker. I gave a talk on analyzing data for intelligent systems using pattern mining techniques. There was also an interesting keynote talk by Prof. Yang Xiao from University of Alabama, USA about detecting the theft of electricity from electricity networks and smart grids. Another keynote speaker was Prof. En Zhu from the National University of Defense Technology, who talked about detecting flow and anomalies in images. The fourth keynote speaker was Prof. Yong Wang from Central South University, about optimization algorithms and edge computing. That presentation has shown some cool applications such as drones being used to improve the internet coverage in some area or optimizing the placement of wind turbines in a wind farm. The last keynote speaker was Prof. Jian Yao from Wuhan University, about image-fusion. He shown many advanced techniques to transform images such as to fix light and stitching together overlaping videos.

This my pass, and program book:

Below, is the registration desk. The staff has been very helpful through the conference:

This is one of the room for listening to the talks:

This is a group picture:

There was also social activities such as an evening dinner and banquet, where I met many interesting researchers that I will keep contact with.

That is all of what I will write for today. It is just to give a quick overview of the conference. Next month, I will write about the ICSI 2021, CCF-AI 2021 , DSIT 2021 , and IEA AIE 2021 conferences, that I will also attend.

—
Philippe Fournier-Viger is a professor of Computer Science and also the founder of the open-source data mining software SPMF, offering more than 120 data mining algorithms.

Posted in Academia, artificial intelligence, Conference, Machine Learning | Tagged changsha, china, computer vision, conference, icivis, image, machine learning | Leave a comment

Key Papers about Episode Mining

What is Machine Learning?

Why Journal Special Issues are Popular?

Useful Latex tricks for Writing Research Papers

Brief report about the IEA AIE 2021 conference

Brief report about the DSIT 2021 conference (4th Intern. Conf. on Data Science and Information Technology)

Brief report about the CCF-AI 2021 conference

Brief report about ICSI 2021 (12th Int. Conference on Swarm Intelligence)

SPMF 2.48

Brief report about ICIVIS 2021 (Int. Conference on Image, Vision and Intelligent system)

Archives

Categories

Recent Posts

Recent Comments

Number of visitors:

Related posts:

Related posts:

Related posts:

Related posts:

Related posts:

Related posts:

Related posts:

Related posts:

Related posts:

Related posts:

Archives

Categories

Recent Posts

Recent Comments

Tag cloud

Number of visitors: