Hi all, This is the call for papers of ICGEC 2021. I am co-organizing a special session there. Welcome to submit your papers! Philippe ===================== The 14th International Conference on Genetic and Evolutionary Computing (ICGEC 2021), is technically co-sponsored by Northeast Electric Power University in China, Fujian University of Technology in China, Shandong University of
Forum: The Data Mining / Big Data Forum

Hi, Just curious, have you succeeded to finish your implementation and get feedback from authors? Best regards, Philippe
Forum: The Data Mining / Big Data Forum

Hi all, Just to let you know that there is a special issue on deep learning for NLP in the Array journal : https://www.journals.elsevier.com/array/call-for-papers/deep-learning-for-natural-language-processing Philippe
Forum: The Data Mining / Big Data Forum

Hi all, I am the guest editor of a special issue in the WCMC journal. Please consider submitting a paper to our special issue: https://www.hindawi.com/journals/wcmc/si/710287/?utm_source=MarketingCloud&utm_medium=email&utm_content=EF_GET_SI2_Engage_E2_Launch_2020-Jan_2021-03-16&utm_campaign=HDW_MRKT_GBL_AWA_EMIL_OWN_GETM_SPEC_X2_10084&utm_term=%%SpecialIssueName%%&ema
Forum: The Data Mining / Big Data Forum

Hi, Have you tried lowering the parameters ? (minimum support and minimum confidence?) The parameter like minimum support must be set to very different values for each datasets. For example, on some datasets, there might be 0 itemsets for minsup =0.4 while for another dataset, there might be millions of itemsets for minsup =0.9999 The best is to start with a high value and gradually de
Forum: The Data Mining / Big Data Forum

Hi, I will fill it now. Best regards, Philippe
Forum: The Data Mining / Big Data Forum

Hi all, This is to let you know that a new version of SPMF is out (v. 2.45) . There are four new algorithms: - CLH-Miner for mining cross-level high utility itemsets (thanks to Bay Vo et al. for the efficient implementation) - FHUQI-Miner, state-of-the-art algorithm for mining high utility quantitative itemsets (by Mourad Nouioua et al.) - POERM and POERM-ALL algorithms for mining partia
Forum: The Data Mining / Big Data Forum

Hi, That is nice. No worry. What is your research area? Are you doing the Phd? Data mining is about using computer software to analyze data to find some useful information in the data. For example, let's say that there is a retail store collecting data about what customers buy. By analyzing this data using software, we can maybe find some interesting information such that many people by
Forum: The Data Mining / Big Data Forum

Hi Welcome to the forum! Yes, it is very old school ;-) I think there are more people reading than posting. Welcome to post. I personally check the forum quite frequently and try to answer questions when I can. Best, Philippe
Forum: The Data Mining / Big Data Forum

Hi all, We have recently published a paper about using artificial intelligence and data mining techniques to analyze the COVID19 genome of different strains. We have used techniques from SPMF such as sequential pattern mining, sequence predictions models, itemset mining, and some new algorithm for mutation analysis. PAPER: http://www.philippe-fournier-viger.com/2021_APIN_Using_Artificial_
Forum: The Data Mining / Big Data Forum

Hi all, For all of those using CPT, CPT+, AKOM, PPM, DG and other sequence prediction models of SPMF, in the new version 2.44, I have added some example in the code about how to save a pretrained model to a file and then load it again. It is very simple. For example, in MainTestCPT.java, the code for saving a trained CPT model and loading it from file to do a prediction looks like this: // *
Forum: The Data Mining / Big Data Forum

Hi, Glad you have found the solution and that the software is useful. Yes, indeed there is an optional parameter. You have found it. Maybe it is not in the documentation? I will check it and update the documentation if needed. I realize that some parameters are maybe not explained. Best regards, Philippe
Forum: The Data Mining / Big Data Forum

Hi all, This is to let you know that I have uploaded the new version 2.44 of SPMF. It has several new algorithms: - LTHUI-Miner for mining the locally-trending high utility itemsets (by Yanjun Yang) - MLHUI-Miner for discovering the multi-level high utility itemsets (by Ying Wang) - AER-Miner for mining attribute evolution rules in a dynamic attributed graph (by Ganghuan He) - TSPIN for mi
Forum: The Data Mining / Big Data Forum

Hello all, I am co-organizing a special issue. See below! =========================================================== Generative Adversarial Networks for Multi-Modal Multimedia Computing Call for papers This Issue is now open for submissions. Papers are published upon acceptance, regardless of the Special Issue publication date. Description Presentation mode and information richn
Forum: The Data Mining / Big Data Forum

Hi, Welcome to the forum! Here is the powerpoint presentation about TKG: http://www.philippe-fournier-viger.com/TKG_frequent_subgraph_mining.pdf The article: http://www.philippe-fournier-viger.com/2019_BDA_TKG_Top-k-subgraphs.pdf To understand the basic idea about TKG, it is good to know first about how gSpan works, since TKG is an extension of GSpan. For this, I recommend to rea
Forum: The Data Mining / Big Data Forum

Great!
Forum: The Data Mining / Big Data Forum

Hi all, This is to let you know that I have uploaded a dozen of videos explaining pattern mining algorithms to my new Youtube channel: https://www.youtube.com/channel/UCk26EiKTBxk1NAQniOV_oyQ/videos?view=0&sort=dd&shelf_id=1 If you are new to pattern mining, there is a lot to learn from these videos. I explain some classical algorithms like Apriori but also some of my own algorith
Forum: The Data Mining / Big Data Forum

Hi again, Yes, that features is not mentioned in the documentation. I forgot to add it. I will do later. There are two optional parameters: - the maximum antecedent size (e.g 2 items) - the maximum consequent size (e.g 3 items) You may add these parameters at the end of the command line like this: java -jar spmf.jar run FPGrowth_association_rules transaction.txt output.txt 0.1% 60%
Forum: The Data Mining / Big Data Forum

Actually, I see that you are using the FPGROWTH version for association rule mining... So you can also try to increase the minconf parameter, and set the constraints on the maximum size of antecedent and consequent of rules! This will help.
Forum: The Data Mining / Big Data Forum

Hi! Thanks for using SPMF. This means that the algorithm is running out of memory. To avoid this problem you may try increasing the "minsup" parameter of FPGrowth, or using the maximum pattern length constraint (see the documentation). This will reduce the number of possibilities and the algorithm will run faster, use less memory and find less patterns. The problem is that in item
Forum: The Data Mining / Big Data Forum

Hi, Yes, I think it is more useful to find periodic patterns that are periodic in each sequence. Like in this paper: http://www.philippe-fournier-viger.com/2019_IS_periodic%20patterns%20multiple%20sequences.pdf If you look at Fig. 1, I think that doing like in Fig. 1 (b) is better than doing like in Fig. 1 (a). I think it would have more applications like for shopping where you have mu
Forum: The Data Mining / Big Data Forum

Hi, Good. I think it could make a paper. But to the important is to write some good motivation for the problem that you are proposing. You should try to find some scenario like about analyzing customer data or analyzing text and explain why finding these periodic sequential rules is interesting. In the introduction of your paper, you can say other algorithms like MPFPS can find periodic patte
Forum: The Data Mining / Big Data Forum

Hi Saed, I did not updated for a while as I was busy. I have added a few more conferences this morning that are coming soon. I will continue updating later. Best regards, Philippe
Forum: The Data Mining / Big Data Forum

Hi all, Just to let you know that there was a small problem in the Ecommerce dataset. It was reported that some transactions contained the same item twice (which is not allowed). I have fixed the problem and re-uploaded these datasets. By the way, I am still working on the next version of SPMF to be released soon. However, I found some bug in some algorithm that I wanted to release and I wan
Forum: The Data Mining / Big Data Forum

Hi, Thanks for letting me know about this problem. I have received that dataset in 2016, and I do not have the file on my computer anymore, nor on my previous laptop it seems. I will try to see if I can find it on my back up hard drives at home later. I will let you know if I can find the original file. Best regards,
Forum: The Data Mining / Big Data Forum

Hi, Thanks for your question. I think you could start from the most simple case which is the frequent periodic sequential rules, and you can think about the case of rare rules as an extension. I think both could be interesting. The main steps would be to define the problem clearly, define an algorithm, and then try it on some idea and look at the patterns that you have found to see what kind
Forum: The Data Mining / Big Data Forum

Hi The item numbers seems to carry some information. For example, i see quickly several types of eggs are in the range 5000-5005 so there seems to be some meaning to these IDs. However, we don't have information about their meaning so I am not sure that we can do something useful from that. But, we do have the taxonomy of items, in the file that is on the dataset page of SPMF. Best regard
Forum: The Data Mining / Big Data Forum

Wish you all a merry Christmas and happy new year! Thanks for your support for the SPMF software and for using/reading the forum
Forum: The Data Mining / Big Data Forum

Enrique Wrote: ------------------------------------------------------- > Hi all, > > I am a researcher in another field (computational > fluid dynamics), but in a certain project in which > I participate I need to find repeated patterns in > a given sequence. My data structure is as follows: > given a certain alphabet (e.g. {A, B, C, D, E}), I > have a SINGLE seq
Forum: The Data Mining / Big Data Forum

Dear Deniz, Thanks for your message. There is not really a standard way to convert the data. The reason is that there are a wide variety of data depending on the applications. So how to encode your data in a way that is meaningful for your application depends on what you want to do. If your data is a text for example, you may encode each sentence as a sequence of items (where items a
Forum: The Data Mining / Big Data Forum