Jaringan syaraf tiruan adalah model komputasi yang terinspirasi oleh sel syaraf biologis
otak manusia. Jaringan syaraf tiruan menyerupai otak manusia dalam dua hal, yaitu
pengetahuan diperoleh jaringan melalui proses belajar, kekuatan hubungan antar sel
syaraf (neuron) yang dikenal sebagai bobot-bobot sinaptik digunakan untuk
menyimpan pengetahuan. Pada penelitian terdahulu telah dilakukan pengujian 12
algoritma pelatihan jaringan backpropogation untuk menentukan algoritma pelatihan
yang paling cepat. Parameter jaringan yang digunakan adalah maksimum epoch =
10000 (104), learning rate = 0.01, 0.05, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9,
1.0, target error = 0.001 (10-3), dan menggunakan model neuron 10-18-1. Berdasarkan
uji ANOVA dengan tingkat alpha (α) = 5%, algoritma pelatihan Gradient Descent with
Adaptive Learning Rate (traingda) adalah algoritma pelatihan yang paling
cepat dengan rata-rata waktu 0.007485 ± 0.0004782 pada learning rate = 0.9. Pada
penelitian ini dilakukan pengujian 12 algoritma pelatihan untuk mengetahui algoritma
pelatihan yang optimal pada model neuron 5-8-1 dan 5-10-1. Nilai learning rate yang
digunakan adalah 0.01, 0.05, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.0. Berdasarkan
uji ANOVA dengan tingkat alpa (α) = 5% pada model neuron 5-8-1 diperoleh
algoritma yang optimal yaitu algoritma pelatihan Levenberg-
Marquardt (trainlm) dengan rata-rata kecepatan = 0.00664500 pada learning rate
=1.0. Sementara itu pada model neuron 5-10-1 diperoleh algoritma yang optimal yaitu
algoritma pelatihan Levenberg-Marquardt (trainlm) dengan rata-rata kecepatan
= 0.00666000 pada learning rate = 0.3.
Artificial neural networks are computational models inspired by the biological nerve cells of the
human brain. Artificial neural networks resemble the human brain in two ways, namely knowledge
obtained by the network through the learning process and the strength of the relationship among
nerve cells (neurons) known as synaptic weights to store knowledge. Previous research has tested 12
backpropogation network training algorithms to determine the fastest training algorithm. The
network parameters used are maximum epoch = 10000 (104), learning rate = 0.01, 0.05, 0.1, 0.2, 0.3,
0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.0, target error = 0.001 (10-3). It also used the neuron model of 10-18-1.
Based on ANOVA test with an alpha level (α) = 5%, the Gradient Descent with Adaptive Learning Rate
(traingda) training algorithm is the fastest training algorithm with an average time of 0.007485 ±
0.0004782 at a learning rate = 0.9. In this study, testing of 12 training algorithms was carried out to
find out the optimal training algorithm in the 5-8-1 and 5-10-1 neuron models. The learning rate
values are 0.01, 0.05, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.0. Based on the ANOVA test with the
level of absent (α) = 5% in the 5-8-1 neuron model, the optimal algorithm is the Levenberg-
Marquardt (trainlm) training algorithm with an average speed = 0.00664500 at the learning rate =
1.0. Meanwhile in the 5-10-1 neuron model, the optimal algorithm is the Levenberg-Marquardt
(trainlm) algorithm training with an average speed = 0.00666000 at the learning rate = 0.3.
The background of this study is the students’ low caring attitude to the the environment and
learning achievement. This study aimed to improve the attitude of caring for the environment and
learning achievement through the Problem Based Learning (PBL) model using mysterious media
boxes and cards (kokami). This classroom action research was carried out in two cycles, each cycle
consisted of two meetings with one learning plan in each meeting. This research consisted of
planning, action, observatio, and reflection. The subjects of this study were fourth grade students,
with 31 students, consisting of 19 male students and 12 female students. Evaluation tests, teacher
and student activity sheets, and students' environmental care scale sheets were used to collect the
data. The results of improvements can be seen in each cycle. Learners became more interested and
more active, because learning was carried out by a new learning model and kokami as the media.
Learners become more concerned with the classroom environment. It can be seen from the clean
floor, free-waste desk drawer and the well-cared plants in front of the class. Increasing the attitude
of environmental care possessed by students is not only seen from the cleanliness of the classroom
environment but also from the results of the questionnaire. In cycle I, the score of the students'
environmental care attitudes was 2.5 with a percentage of 62.5% in quite good criteria. In cycle II,
the average score increased to 3.5 and with a percentage of 88.1% in very well criteria. Cleanliness
of the classroom environment also affects students' learning achievement, as evidenced by the
increase in the results of evaluation tests. The results of the evaluation test show an increase in
student learning achievement from cycle I to cycle II. The average of science learning achievements
in the first cycle was 67.48 with a completeness percentage of 63.69%, the average score of Bahasa
Indonesia was 66.26 with a completeness percentage of 58.35%. In the second cycle, the average of
science learning achievements was 80.6 with a completeness percentage of 88.3%, the average
score of Bahasa Indonesia was 75.33 with a completeness percentage of 81.66%. The results of this
study indicate that the use of the Problem Based Learning (PBL) model with mysterious card box
(kokami) as the media can improve students’ environmental attitudes and learning achievement of
in theme 9, “Kayanya Negeriku” (How Wealth My Country is) for the fourth grade of SD Negeri 2
Kebasen, Kebasen, Banyumas.
Keywords: environmental care attitude, learning achievement, and problem based learning (PBL)
models
The teacher's teaching style is very important in learning, because one of the external
factors that affects students’ learning achievement is the teacher's teaching method or style. This
study aimed to analyze the teaching style of exellent teachers in teaching-learning process at the
Kembaran Subdistrict. This study includes the teacher's constraints in using certain forms of teaching
style and problem solutions due to the teaching style. This study is a qualitative research study on
the subject of Physical Education teachers and students of SD Negeri Larangan, the first grade
teachers and students at SD N 1 Bantarwuni, and fourth grade teachers and the students at SD N
Karangsari. Data were obtained through interviews, observations, and documentation. Test the
validity of the data used technical triangulation and source triangulation. The data obtained were
then analyzed using the Miles and Huberman analysis model to obtain conclusions. The results show
that Physical Education teachers at SDN Larangan tend to use the form of interactional teaching
style, the first grade teachers at SD N 1 Bantarwuni and the fifth grade teachers at SD N Karangsari
tend to use forms of personalization teaching style. It was also found some constraints in applying
the form of a classical teaching style namely students were easily bored and did not pay attention to
the teacher when the teacher explained the subject matter. Meanwhile, the constraints in applying
the form of technological teaching style are the limited facilities and infrastructure. The constraints
in applying the form of personalization teaching style was its difficulties to students’ condition. The
obstacle in applying form of interactional teaching style was the presence of passive students. The
solution to the constraints in doing classical teaching style is that the teacher always gives eye
contact to students, the teacher must pay attention to face expressions and gestures. The obstacles
in doing technological teaching style can be done by involving the environment as a source of
learning. The constraints in doing personalization teaching style can be solved by the use of peers'
methods and giving jokes. Meanwhile, the obstacles in conducting interactional teaching styles can
be solved by appointing passive students to come forward and giving students the opportunity to
ask questions and express opinions.