An Experimental Comparison of Kernel Clustering Methods

نویسندگان

  • Maurizio Filippone
  • Francesco Masulli
  • Stefano Rovetta
چکیده

In this paper, we compare the performances of some among the most popular kernel clustering methods on several data sets. The methods are all based on central clustering and incorporate in various ways the concepts of fuzzy clustering and kernel machines. The data sets are a sample of several application domains and sizes. A thorough discussion about the techniques for validating results is also presented. Results indicate that clustering in kernel space generally outperforms standard clustering, although no method can be proven to be consistently better than the others.

برای دانلود رایگان متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

یادگیری نیمه نظارتی کرنل مرکب با استفاده از تکنیک‌های یادگیری معیار فاصله

Distance metric has a key role in many machine learning and computer vision algorithms so that choosing an appropriate distance metric has a direct effect on the performance of such algorithms. Recently, distance metric learning using labeled data or other available supervisory information has become a very active research area in machine learning applications. Studies in this area have shown t...

متن کامل

Central Clustering in Kernel-induced Spaces Central Clustering in Kernel-induced Spaces Title: Central Clustering in Kernel-induced Spaces

Clustering is the problem of grouping objects on the basis of a similarity measure. Clustering algorithms are a class of useful tools to explore structures in data. Nowadays, the size of data collections is steadily increasing, due to high throughput measurement systems and mass production of information. This makes human intervention and analysis unmanageable without the aid of automatic and u...

متن کامل

Composite Kernel Optimization in Semi-Supervised Metric

Machine-learning solutions to classification, clustering and matching problems critically depend on the adopted metric, which in the past was selected heuristically. In the last decade, it has been demonstrated that an appropriate metric can be learnt from data, resulting in superior performance as compared with traditional metrics. This has recently stimulated a considerable interest in the to...

متن کامل

Max-margin Clustering: Detecting Margins from Projections of Points on Lines - Supplementary material

We provide the results of clustering accuracy for our method across different kernel parameter sets, for cases where the exact number of clusters k is unknown in Table 1. We used a range of values of k from the minimum to maximum number of clusters in those datasets. For synthetic data, 2 ≤ k ≤ 10, for comparison with max-margin clustering methods [1], 2 ≤ k ≤ 9, and for comparison with discrim...

متن کامل

An interior-point algorithm for $P_{ast}(kappa)$-linear complementarity problem based on a new trigonometric kernel function

In this paper, an interior-point algorithm  for $P_{ast}(kappa)$-Linear Complementarity Problem (LCP) based on a new parametric trigonometric kernel function is proposed. By applying strictly feasible starting point condition and using some simple analysis tools, we prove that our algorithm has $O((1+2kappa)sqrt{n} log nlogfrac{n}{epsilon})$ iteration bound for large-update methods, which coinc...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2008