No abstract

Identifying beef manually has some drawbacks because human visual has limitations and there are differences of human perception in assessing object quality. Several researches developed beef quality assessment methods based on image feature extraction. However, not all features support for obtaining the classification results that have high accuracy. The efficiency will be achieved if the classification analyzes only the relevant features. Therefore, a feature selection process is required to select relevant features and to eliminate irrelevant features to obtain more accurate and faster classification results. One of the feature selection algorithms is the F-Score which is a simple technique that measures the discrimination of two sets of real numbers. The features with the lowest ranking from the F-Score will be eliminated one by one until the most relevant features are obtained. The test is carried out by analyzing the classification results in the form of sensitivity, specificity, and accuracy values. The results of this research showed that by using the F-Score feature, the most relevant features for the classification of freshness level of local beef are obtained using the K-Nearest Neighbor (KNN) method. These features include the average color intensity R and standard deviation with a sensitivity of 0.8, a specificity of 0.93, and an accuracy of 86%.  Keywords:  Classification, Fiture Selection, F-Score, K-Nearest Neighbor, Local beef

Perkembangan Gigabit Ethernet mengalami peningkatan yang cukup pesat. Di Universitas Lampung sendiri telah menggunakan teknologi Gigabit Ethernet sebagai backbone karena mampu mentransmisikan data yang besar serta berkecepatan tinggi. Dengan adanya teknologi Gigabit Ethernet ini, diharapkan peningkatan kualitas intranet yang ada di Universitas Lampung semakin baik kedepannya.Implementasi dari jaringan intranet dengan teknologi Gigabit Ethernet ini perlu diketahui kinerjanya. Pada penelitian ini digunakan metode Design Science Research (DSR) yang memiliki 6 tahapan yaitu identifikasi masalah dan motivasi, menetapkan objek solusi, desain dan pengembangan, demonstrasi, evaluasi, dan pelaporan hasil. Analisa ilmiah ini dilakukan dengan pengukuran terhadap trafik dari jaringan Local Area Netwok (LAN). Parameter yang diukur dan dianalisa adalah bandwidth, delay, jitter, dan packet loss dengan pemberian beban berupa paket data pada TCP dan UDP untuk melihat karakteristik dari jaringan tersebut dengan menggunakan software Jperf dan Wireshark guna melihat baik atau buruknya kualitas dari jaringan intranet Unila. Hasil dari penelitian ini yaitu dapat dikategorikan “Baik” berdasarkan acuan standar ITU-T G.114.Kata kunci: Jperf, Wireshark, Quality of Service, Gigabit Ethernet, DSR, ITU-T G.114.

Abstrak. Pendaftaran seminar masih harus menemui admin di jurusan pada jam kerja sementara itu terkadang admin tidak ada ditempat, selain itu admin harus mengirimkan persyaratan seminar berulang kepada mahasiswa yang ingin mendaftar. Dalam penelitian ini akan dikembangkan modul pendaftaran seminar untuk mengintegrasikan  layanan-layanan informasi di program studi. Penelitian ini bertujuan mengembangkan Sistem Informasi Portal Prodi Modul Pendaftaran Seminar Akademik Di Jurusan Teknik Elektro Fakultas Teknik Universitas Lampung Menggunakan Metode Rapid Application Development (RAD). Hasil pengujian fungsi menggunakan metode Black Box Testing  memperoleh hasil dalam keseluruhan pengujian pada fungsi mendapatkan hasil yaitu OK/berhasil tanpa adanya FAIL/kegagalan pada saat pengujian fungsi. Pengujian terhadap pengguna pada sistem informasi portal prodi modul pendaftaran seminar akademik menggunakan metode pengujian User Experience Question (UEQ) yaitu pada aspek Attractiveness, Perspicuity, Efficiency, Dependability, Stimulation dan Novelty dengan jumlah responden 30 orang. Berdasarkan pengujian tersebut, didapatkan penilaian pengujian pada aspek Attractiveness mendapatkan nilai 1.84 (Excellent),  Perspicuity mendapatkan nilai 1.93 (Good), Efficiency mendapatkan nilai 1.71 (Good), Dependability nilai 1,56 (Good) , Stimulation mendapatkan nilai 1.88 (Excellent) dan Novelty mendapatkan nilai 1.36 (Good). Hasil ini termasuk kedalam kategori  Acceptable/good melihat dari hasil aspek pengujian UEQ.

Abstract. Hoax news is fake news that contains information that intentionally misleads people and has a specific political agenda. Along with the development of technological developments, the news is increasingly unclear whether the truth is only in accordance with facts or mere hoaxes. Based on this problem, hoax news is carried out, one of which uses a machine that can process news classification automatically. Machine learning is implemented using a framework called the Flask framework that can run on both on-premises servers and cloud computing. Local servers with computational limitations that are static in nature have problems running large computations on the hoax news classification system framework which is characterized by a predictable time schedule so that a distributed automatic-scaling computing system can handle cloud computing loads. Cloud computing offers compute load-sharing automatic scalability that can provide stable compute computing. So this research focuses on the application of the machine learning hoax news classification model into the framework of the machine learning deployment Platform as a Service (PaaS) model in Cloud computing called Google App Engine (GAE). The application of the hoax news classification system in the Google App Engine environment runs with an average prediction time of 11.53 seconds, better and more stable than the local server's average prediction of 17.50 seconds.