Pengembangan Alat Evaluasi melalui Aplikasi Kahoot Materi Karaktekteristik Ruang dan Pemanfaatan Sumber Daya Alam dengan Penguatan Karakter Mandiri Kelas IV Mi.Rn

Research on the development of evaluation tools kahoot the application of material space spatial characteristics and the utilization of natural resource themes with the Strengthening of Independent Character class IV MI.RN are valid according to material experts, media experts, as well as practical for teachers and students. The method used is a 10-step model from Borg and Gall. Evaluation tools through the application of kahoot have produced a percentage value of 93.33 percent material experts, 100 percent media experts and 97.5 percent user (teacher). Abstrak: Penelitian pengembangan alat evaluasi melalui aplikasi kahoot materi karakteristik ruang dan pemanfaatan sumber daya alam tema dengan Penguatan Karakter Mandiri kelas IV MI.RN yang valid menurut ahli materi, ahli media, serta praktis bagi guru dan siswa. Metode yang digunakan adalah 10 langkah model dari Borg and Gall. Alat evaluasi melalui apikasi kahoot telah menghasilkan nilai persentase dari ahli materi 93,33 persen, ahli media 100 persen dan pengguna (guru) 97,5 persen.

TextScanner: Reading Characters in Order for Robust Scene Text Recognition

Driven by deep learning and a large volume of data, scene text recognition has evolved rapidly in recent years. Formerly, RNN-attention-based methods have dominated this field, but suffer from the problem of attention drift in certain situations. Lately, semantic segmentation based algorithms have proven effective at recognizing text of different forms (horizontal, oriented and curved). However, these methods may produce spurious characters or miss genuine characters, as they rely heavily on a thresholding procedure operated on segmentation maps. To tackle these challenges, we propose in this paper an alternative approach, called TextScanner, for scene text recognition. TextScanner bears three characteristics: (1) Basically, it belongs to the semantic segmentation family, as it generates pixel-wise, multi-channel segmentation maps for character class, position and order; (2) Meanwhile, akin to RNN-attention-based methods, it also adopts RNN for context modeling; (3) Moreover, it performs paralleled prediction for character position and class, and ensures that characters are transcripted in the correct order. The experiments on standard benchmark datasets demonstrate that TextScanner outperforms the state-of-the-art methods. Moreover, TextScanner shows its superiority in recognizing more difficult text such as Chinese transcripts and aligning with target characters.

Pengembangan Buku Ajar Menggunakan Model 4D dalam Peningkatan Keberhasilan Pembelajaran Pendidikan Agama Islam

It is a fact that, teachers have been accustomed to take materials as teaching materials from textbooks or textbooks provided the school and though the textbook still has many weaknesses. Among them are: (1) No Standard Competency, Basic Competency, and Indicator, (2) there are no instructional guidelines, (3) no multiple choice assessment, (4) lack of learning material on reference of reading material and (5) unsuitable laying of the Qur'an and hadith. From some of the above problems, the researcher is interested to write a scientific paper entitled Development of Textbook Using 4 D Model (Define, Design, Develop, And Disseminate) in improving success of learning in islamic studies lesson class x even semester in SMA Ibrahimy Sukorejo Situbondo on 2016/2017 year. The research method used is R & D Using Model 4 D (Define, Design, Develop, And Disseminate). Technique of data collection using test method, questionnaire, observation, interview, and documentation. while the data analysis using the test device analysis and effectiveness test. The development has resulted in one textbook product, entitled "Textbook of Islamic Education And Nature Character Class X in second Semester.”

EXPERIMENTAL ANALYSIS OF INDIVIDUALITY OF JAPANESE HANDWRITING

A handwriting examination to determine the writer (writer verification and writer identification) is based upon the hypothesis that each person has characteristic handwriting that is distinct from the handwriting of other individuals. This hypothesis has been accepted empirically through many case studies. However, this hypothesis must be established with scientific and quantitative rigor in order to increase its admissibility as forensic evidence. This paper describes the results of a quantitative investigation into individuality of Japanese handwriting. The authors investigated the individuality (consistency and discriminability) of normal Japanese handwriting using variation entropy and writer verification algorithms. We also compared these results with the analysis results of Japanese signatures that we investigated previously. The authors discuss the differences of individuality between normal handwriting and signatures. Individuality of normal handwriting becomes noticeable as the number of strokes increases, and varies with the complexity of the character class. Although individuality of both normal handwriting and signatures also depends on the writer, this is more noticeable for signatures. It is important that these facts are considered in actual forensic analysis of Japanese handwriting.