Aboul Ella Hassanien (Abo)

In this paper, a system for feature selection based
on chicken swarm optimization (CSO) algorithm is proposed.
Datasets ordinarily includes a huge number of attributes, with
irrelevant and redundant attribute. Commonly wrapper-based
approaches are used for feature selection but it always requires
an intelligent search technique as part of the evaluation function.
Chicken swarm optimization (CSO)is a new bio-inspired
algorithm mimicking the hierarchal order of the chicken
swarm and the behaviors of chicken swarm, including roosters,
hens and chicks, CSO can efficiently extract the chickens’
swarm intelligence to optimize problems. Therefore, CSO was
employed to feature selection in wrapper mode to search
the feature space for optimal feature combination maximizing
classification performance, while minimizing the number of
selected features. The proposed system was benchmarked
on 18 datasets drawn from the UCI repository and using
different evaluation criteria and proves advance over particle
swarm optimization (PSO) and genetic algorithms (GA) that
commonly used in optimization problems

In this paper, a proposed system for feature selection
based on social spider optimization (SSO) is proposed. SSO is
used in the proposed system as searching method to find optimal
feature set maximizing classification performance and mimics
the cooperative behavior mechanism of social spiders in nature.
The proposed SSO algorithm considers two different search
agents (social members) male and female spiders, that simulate
a group of spiders with interaction to each other based on the
biological laws of the cooperative colony. Depending on spider
gender, each spider (individual) is simulating a set of different
evolutionary operators of different cooperative behaviors that are
typically found in the colony. The proposed system is evaluated
using different evaluation criteria on 18 different datasets, which
compared with two common search methods namely particle
swarm optimization (PSO), and genetic algorithm (GA). SSO
algorithm proves an advance in classification performance using
different evaluation indicators

Multidimensionality in social networks is a great issue that
came out into view as a result of that most social media sites such as
Facebook, Twitter, and YouTube allow people to interact with each other
through dierent social activities. The community detection in such mul-
tidimensional social networks has attracted a lot of attention in the recent
years. When dealing with these networks the concept of community de-
tection changes to be, the discovery of the shared group structure across
all network dimensions such that members in the same group interact
with each other more frequently than those outside the group. Most of
the studies presented on the topic of community detection assume that
there is only one kind of relation in the network. In this paper, we propose
a multi-objective approach, named MOGA-MDNet, to discover commu-
nities in multidimensional networks, by applying genetic algorithms. The
method aims to nd community structure that simultaneously maximizes
modularity, as an objective function, in all network dimensions. This
method does not need any prior knowledge about number of communi-
ties. Experiments on synthetic and real life networks show the capability
of the proposed algorithm to successfully detect the structure hidden
within these networks.

This paper presents Case-Based Reasoning (CBR) system to asses water pollution based on fish liver histopathology as biomarker. The proposed approach utilizes fish liver microscopic images in order to asses water pollution based on knowledge stored in the case-based database and stores likelihood description of the previous solutions in order to make the knowledge stored more flexible. The proposed case-based reasoning system consists of 5 phases; namely case representation (pre-processing and feature extraction), retrieve, reuse/adapt, revise, and retain phases. After applying pre-processing and feature extraction algorithms on the input images, similarity between the input and case base database is being calculated in order to retrieve similarity. Experimental results show that the performance of CBR systems increases according to the number of retrieved cases in each scenario against each strategy. The proposed system achieved 95.9

Disabilities, specially Spinal Cord Injuries (SCI), affect people behaviors, their response, and the participation in daily activities. People with SCI need long care, cost, and time to improve their heath status. So, the rehabilitation of people with SCI on different period of times is required. In this paper, we proposed an automated system to estimate the rehabilitation length of stay of patients with SCI. The proposed system is divided into three phases; (1) pre-processing phase, (2) classification phase, and (3) rehabilitation length of stay measurement phase. The proposed system is automating International Classification of Functioning, Disability and Health classification (ICF) coding process, monitoring progress in patient status, and measuring the rehabilitation time based on support vector machines algorithm. The proposed system used linear and radial basis (RBF) kernel functions of support vector machines (SVMs) classification algorithm to classify data. The accuracy obtained was full match on training and testing data for linear kernel function and 93.3 % match for RBF kernel function.

Reality Mining is one of the first efforts that have been exerted to utilize smartphone’s data; to analyze human behavior. The smartphone data are used to identify human behavior and discover more attributes about smartphone users, such as their personality traits and their relationship status. Text messages and SMS logs are two of the main data resources from the smartphones. In this paper, The proposed system define the user personality by observing behavioral characteristics derived from smartphone logs and the language used in text messages. Hence, The supervised machine learning methods (K-nearest nighbor (KNN), support vector machine, and Naive Bayes) and text mining techniques are used in studying the textual matter messages. From this study, The correlation between text messages and predicate users personality traits is broken down. The results provided an overview on how text messages and smartphone logs represent the user behavior; as they chew over the user personality traits with accuracy up to 70 %.

This paper proposes a human thermal face recognition approach with two variants based on Random linear
Oracle (RLO) ensembles. For the two approaches, the Segmentation-based Fractal Texture Analysis (SFTA) algorithm was used for extracting features and the RLO ensemble classifier was used for recognizing the face from its thermal image. For the dimensionality reduction, one variant (SFTALDA-RLO) was used the technique of Linear Discriminant Analysis (LDA) while the other variant (SFTA-PCA-RLO) was used the Principal Component Analysis (PCA). The classifier’s model was built using the RLO classifier during the training phase and in the testing phase then this model was used to identify the unknown sample images. The two variants were evaluated using the Terravic Facial IR Database and the experimental results showed that the two variants achieved a good recognition rate at 94.12% which is better than related work.

Starting from the last century, animals identification
became important for several purposes, e.g. tracking,
controlling livestock transaction, and illness control. Invasive and
traditional ways used to achieve such animal identification in
farms or laboratories. To avoid such invasiveness and to get more
accurate identification results, biometric identification methods
have appeared. This paper presents an invariant biometric-based
identification system to identify cattle based on their muzzle
print images. This system makes use of Speeded Up Robust
Feature (SURF) features extraction technique along with with
minimum distance and Support Vector Machine (SVM) classifiers.
The proposed system targets to get best accuracy using minimum
number of SURF interest points, which minimizes the time
needed for the system to complete an accurate identification.
It also compares between the accuracy gained from SURF
features through different classifiers. The experiments run 217
muzzle print images and the experimental results showed that
our proposed approach achieved an excellent identification rate
compared with other previous works.

This paper intends to enhance the image for the next usage
of region growing technique for segmenting the region of liver away from
other organs. The approach depends on a preprocessing phase to enhance
the appearance of the boundaries of the liver. This is performed using
contrast stretching and some morphological operations to prepare the
image for next segmentation phase. The approach starts with combining
Otsu's global thresholding with dilation and erosion to remove image
annotation and machine's bed. The second step of image preparation
is to connect ribs, and apply lters to enhance image and deepen liver
boundaries. The combined lters are contrast stretching and texture l-
ters. The last step is to use a simple region growing technique, which has
low computational cost, but ignored for its low accuracy. The proposed
approach is appropriate for many images, where liver could not be sep-
arated before, because of the similarity of the intensity with other close
organs. A set of 44 images taken in pre-contrast phase, were used to test
the approach. Validating the approach has been done using similarity
index. The experimental results, show that the overall accuracy oered
by the proposed approach results in 91.3% accuracy.