Graph kernels based on linear patterns: Theoretical and experimental comparisons
نویسندگان
چکیده
Graph kernels are powerful tools to bridge the gap between machine learning and data encoded as graphs. Most graph based on decomposition of graphs into a set patterns. The similarity two is then deduced corresponding Kernels linear patterns constitute good trade-off accuracy computational complexity. In this work, we propose thorough investigation comparison different patterns, namely walks paths. First, all these explored in detail, including their mathematical foundations, structures After that, experiments performed various benchmark datasets exhibiting types graphs, labeled unlabeled with numbers vertices, average vertex degrees, non-linear Finally, for regression classification tasks, complexity compared analyzed, light baseline Suggestions proposed choose according datasets. This work leads clear strengths weaknesses kernels. An open-source Python library containing an implementation discussed publicly available GitHub community, thus allowing promote facilitate use problems.
منابع مشابه
on the effect of linear & non-linear texts on students comprehension and recalling
چکیده ندارد.
15 صفحه اولtask-based language teaching in iran: a mixed study through constructing and validating a new questionnaire based on theoretical, sociocultural, and educational frameworks
جنبه های گوناگونی از زندگی در ایران را از جمله سبک زندگی، علم و امکانات فنی و تکنولوژیکی می توان کم یا بیش وارداتی در نظر گرفت. زبان انگلیسی و روش تدریس آن نیز از این قاعده مثتسنی نیست. با این حال گاهی سوال پیش می آید که آیا یک روش خاص با زیر ساخت های نظری، فرهنگی اجتماعی و آموزشی جامعه ایرانی سازگاری دارد یا خیر. این تحقیق بر اساس روش های ترکیبی انجام شده است.پرسش نامه ای نیز برای زبان آموزان ...
Graph Kernels based on High Order Graphlet Parsing and Hashing
Graph-based methods are known to be successful in many machine learning and pattern classification tasks. These methods consider semi-structured data as graphs where nodes correspond to primitives (parts, interest points, segments, etc.) and edges characterize the relationships between these primitives. However, these non-vectorial graph data cannot be straightforwardly plugged into off-the-she...
متن کاملA theoretical framework for gene induction and experimental comparisons.
Ligand-mediated gene induction by steroid receptors is a multistep process characterized by a dose-response curve for gene product that follows a first-order Hill equation. This behavior has classically been explained by steroid binding to receptor being the rate-limiting step. However, this predicts a constant potency of gene induction (EC(50)) for a given receptor-steroid complex, which is ch...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Expert Systems With Applications
سال: 2022
ISSN: ['1873-6793', '0957-4174']
DOI: https://doi.org/10.1016/j.eswa.2021.116095