Information Sciences

Three kinds of attributed networks are used in our experiments. They are Cora data set, Citeseer data set, and Wiki data set.

Normalized Input and output variable from a nonlinear dynamical MIMO SRU process of a refinery.

This is includes both datasets and the source code of algorithms implemented in Matlab.

We simulate the attack scenario on the NS-2 emulator. The network consists of three routers, 25 users (10 users for background traffic) and one attacker. The bottleneck link is from router 2 to router 3 with a bandwidth of 10 Mbps. The users access the server by establishing TCP connections. The attacker sends UDP packets for low-rate DoS attack. The TestBed is a real network scenario we build with seven hosts (6 users and 1 attacker) and two routers. The bottleneck link is from router 1 to router 2 with a bandwidth of 10 Mbps.

The file contains 18 datasets, including 12 UCI synthetic datasets and 6 real datasets. The synthetic data set includes spherical data, strip data, irregular distribution data, composite data, unbalanced data, and the data of random addition of outliers on its basis. The real data set has been normalized and can be directly used for processing and analysis.

Data for: An improved multi-stage genetic programming model for streamflow prediction Monthly streamflow data of the Sedre River during the 1987–2015 period

The raw data of the Hip-DE algorithm. Code can be fetched via: https://sites.google.com/view/zhenyumeng after acceptance of our paper.

An example of target recognition using ESM data, to compare the classic Bayesian classifier with a fuzzy classifier with heterogeneous information fusion.

This repo contains 42 highly imbalanced datasets for the paper "A Hybrid Data-Level Ensemble (HD-Ensemble) for Highly Imbalance Learning". 28 of the datasets were extracted from KEEL, and 14 were manually created by ourselvies.

We provide the source code of this paper for reproducibility.