Major conferences in the field of data mining [reprinted]Http://blogger.org.cn/blog/more.asp? Name = zhaoyong04 & id = 24556
First-class: sigmod, vldb, icde, data mining KDD, machine learning icml, SIGIR for information retrieval, and pods for database theory meetings, but it is a theoretical meeting, so it is not relevant to us.
Dual-stream: edbt, ICDT, cikm, SDM, ICDM, pkdd, and ecml European machine learning conference (this should be the first class, better than the second class ), these will not be good either. If you are in the middle of the game, you may feel that you will be first-class with your efforts, haha. There are some other sessions, so I am too lazy to list them!The following is a post of AI edition which was written quite well. Let's take a look. Haha
------------------------------------ Some people have original jobs and there are always some very new things every year. Some people have a lot of articles, but they mostly share the work of follow. There are many paper machines in the database field. In some cases, the entire group is a large papermachine. I personally feel that database researchers tend to regard data mining as a sub-field of a database.
Low rating. However, for those in other backgrounds, data mining is a relatively independent emerging field.
Rating is relatively high. Sigmod: 97 points, the highest meeting of the database, which involves a wide range of applications (because theoretical articles have pods ). No
That's all. This meeting is not only a double-blind review, but also a rebuttal procedur
E is unique and distinctive. Vldb: 95 points, very good database conference. Similar to sigmod, it has a wide range of applications. In terms of the quality of the article, sigmod and vldb are difficult to distinguish, not to mention who is better than who is. Their scope is almost the same.
Many cool people believe that this year's rebuttal procedure was not very successful. Too many submissions, it is difficult to achieve each
Both are fair and equitable. A lot of rebuttal is not visible. Double-blind is a double-edged sword. Over the past few years, some people have contributed documents in the style of impersonating a cool man.
On the contrary, the quality of vldb reviews has been high. Every year, vldb has a very theoretical paper. In general, I think sigmod is better. Based on the articles I have read
. However, this is not important. The difference is also true. Pods: 95 points. It is the "best meeting on database theory and a good theoretical meeting ". Always co-locate every year
D With sigmod. People with algorithm backgrounds are dominant (you can count the number of people in pods articles from motwani
Group), and some people with AI backgrounds (after all, sigart is also one of the sponsors ). It has far less influence than sIgM
Od, but the quality of the article is relatively neat, variance is less than sigmod (and any other database Conference ). Yes
"Pods never had a re
Ally bad paper, "it's proud. KDD: full paper 95 points, poster/short paper 90 points. The highest meeting for data mining. Due to historical accumulation
Insufficient and the domain circle is small. Do not say that KDD is inferior to sigmod at present. I think we can use this analogy.
: KDD: sigmod = crypto: stoc. Looking back at the history of cryptography, the best article is generally published in stoc/focs rather than C
Rypto/eurocrypt, which is similar to today's data mining! However, if you look at today's cryptographic articles
Top-level cryptographic companies (which cannot be written) will not contribute to stoc/focs any more. I think the same thing will happen soon.
It will also happen in Data Mining in the future. Let's wait and see. In the past few years, KDD has a high quality. Its full paper quality is higher than that of sigmod/vldb in data mining.
Quality. The reason is that there are very few data mining personnel in sigmod/vldb reviewers, and the review criteria may not be well grasped.
Several pieces of sigmod/vldb Data Mining paper have follow some KDD paper over the past few years. In KDD, we need to get an article F.
Ull paper is really difficult. Fudan took an article last year, which is very valuable. This year they took another sigmod demo
Ming's work is indeed solid. I have heard that in many places, if you have a sigmod, vldb, or KDD, You can graduate from a doctor. If you have two, you can find it good.
. "The revolution has not yet succeeded. comrades still need to work hard !" Icde: 92 points. A good Database Conference is also a hodgedge. The advantage is wide coverage and high inclusiveness. The disadvantage is that
Chapter levels are uneven. Edbt: 88 points, a good database conference, the recording rate is very low, but the history accumulation is insufficient, the impact is obviously inferior to icde. ICDT: 88 points, European version of pods, Second Meeting on database theory. Like sigmod/vldb, icde and edbt are comparable in quality and impact.
Other problems, such as cikm, ICDM, SDM, ssdbm, and pkdd, are worse than the preceding ones. Cikm: 85 points. SDM: full paper 90 points, poster/short paper 85 points. Siam's data mining conference ranks second in the field of data mining with ICDM, which is significantly different from KDD. It seems that there are many people with statistical backgrounds, and some people with machine learning backgrounds are compared with diversified. ICDM: full paper 90 points, poster/short paper 85 points. IEEE data mining conference, which is listed with SDM
The second place in the data mining field is significantly different from that in KDD. Pkdd: 83 points (because the number of poster/short papers is small, it is not distinguished ). Like the European version of KDD,
However, there is a big gap with KDD. |
