Hasil Pencarian  ::  Simpan CSV :: Kembali

Abstrak :
Selama pemberlakuan pembatasan kegiatan masyarakat (PPKM) saat pandemi Covid-19, berbagai aktivitas bekerja, belajar, hingga berbelanja lebih banyak dilakukan dengan memanfaatkan teknologi digital dari rumah. Telkomsel Orbit yang merupakan produk layanan internet rumahan hadir untuk mendukung perubahan perilaku pelanggan selama pandemi. Pelanggan Telkomsel Orbit telah meningkat secara eksponensial selama pandemi, akan tetapi pada bulan Oktober 2021 hingga Januari 2022 terjadi penurunan jumlah pertumbuhan pelanggan. Pemasaran di media sosial dengan menggandeng influencer di media sosial dapat dilakukan untuk meningkatkan pertumbuhan pelanggan dan mengatasi masalah tersebut. Penelitian ini dilakukan untuk mendeteksi influencer di media sosial Twitter dan Instagram dengan menggunakan analisis sentimen dan jejaring sosial untuk produk Telkomsel Orbit. Data yang digunakan merupakan tweet, retweet, postingan, dan komentar yang diunggah pada periode 1 Oktober 2021 hingga 31 Maret 2022 terkait produk Telkomsel Orbit, total sebanyak 6,092 tweet dan 8,095 postingan dikumpulkan. Penelitian ini menggunakan algoritma Naïve Bayes (NB), Support Vector Machine (SVM), dan Long Short-Term Memory Recurrent Neural Network (LSTM-RNN) pada proses pembuatan model dalam analisis sentimen berbasis pembelajaran mesin. Selain itu, penelitian ini menggunakan analisis jejaring sosial untuk menentukan aktor utama yang memiliki pengaruh terbesar dengan mengukur empat nilai sentralitas yaitu degree centrality, closeness centrality, betweenness centrality, dan eigenvector centrality. Hasil penelitian menunjukan algoritma SVM memiliki kinerja terbaik dengan rata-rata nilai akurasi sebesar 83.68% diikuti oleh LSTM-RNN sebesar 82.54% dan NB sebesar 75.35%. Selain itu, akun dengan pengaruh terbesar berdasarkan nilai sentralitasnya pada media sosial Twitter adalah denkmit untuk sentimen positif dan myorbitid untuk sentimen negatif, sedangkan pada media sosial Instagram adalah akun tseljabotabekjabar, witelaceh, telkomsel.halo.bjm, dan telkomsel.bojonegoro untuk sentimen positif dan negatif. Hasil dari penelitian ini memberikan kontribusi praktis bagi organisasi dalam melakukan pemasaran di media sosial untuk membangun merek serta menyebarkan informasi dan promosi secara elektronik untuk meningkatkan niat beli suatu produk. Influencer di media sosial berperan penting sebagai pihak yang menyebarkan informasi dan promosi tersebut di media sosial. Selain itu, penelitian ini juga memberikan kontribusi teoritis dalam analisis sentimen berbasis pembelajaran mesin dan analisis jejaring sosial dalam mendeteksi aktor utama di media sosial. ......During the implementation of restrictions on community activities (PPKM) in times of the Covid-19 pandemic, various activities such as working, studying, and shopping were mostly carried out by utilizing digital technology from home. Telkomsel Orbit, a home internet service was launched to support changes in customer behavior during the pandemic. Telkomsel Orbit subscribers have increased exponentially during the pandemic, but from October 2021 to January 2022 there was a decrease in the number of subscriber growth. To overcome these problems and increase customer growth, social media marketing and collaboration with influencers on social media could be leveraged. This research was conducted to detect influencers on social media, namely Twitter and Instagram using sentiment analysis and social networks analysis for Telkomsel Orbit products. The data used were tweets, retweets, posts, and comments uploaded in the period between October 1, 2021 and March 31, 2022 related to Telkomsel Orbit products, a total of 6,092 tweets and 8,095 posts were collected. This study used Naïve Bayes (NB), Support Vector Machine (SVM), and Long Short-Term Memory Recurrent Neural Network (LSTM-RNN) algorithm in machine learning-based sentiment analysis. In addition, social networks analysis was also conducted by measuring four centrality metrics, namely degree centrality, closeness centrality, betweenness centrality, and eigenvector centrality. The results of this study showed that SVM algorithm had the best performance with an average accuracy of 83.68% followed by LSTM-RNN 82.54% and NB 75.35%. In addition, the accounts with the greatest influence based on the metric of centrality on Twitter are denkmit for positive sentiment and myorbitid for negative sentiment, while on Instagram social media are tseljabotabekjabar, witelaceh, telkomsel.halo.bjm, and telkomsel.bojonegoro for both of positive and negative sentiment. The results of this study provide a practical contribution to organizations in marketing on social media to build brands and disseminate information and promotions electronically to increase purchase intention of a product. Influencers on social media play an important role as third parties who will disseminate information and promotions on social media. In addition, this research also provides theoretical contributions in machine learning-based sentiment analysis and social network analysis in detecting the main actors in social media.

Abstrak :
ABSTRAK Rencana Strategis Teknologi Informasi PT XYZ berkomitmen menerapkan sistem pengamanan dan monitoring berupa pengembangan kehandalan keamanan infrastruktur TI. Namun kenyataannya masih ditemukan kejadian insiden keamanan yang mengakibatkan kerugian finansial maupun non-finansial, selain itu hasil pengukuran oleh Hewlet Packard Enterprise (HPE) pada tahun 2016 menyatakan tingkat kematangan security operations PT XYZ berada di level Incomplete, sehingga PT XYZ memiliki permasalahan pengelolaan security operations yang belum layak. Penelitian ini mengevaluasi pengelolaan security operations saat ini berdasarkan hasil pengukuran HPE sebelumnya dan melakukan pengukuran tingkat kematangan secara mandiri menggunakan tools SOC-CMM. Perancangan Security Operations Center (SOC) menggunakan pendekatan Design Science Research (DSR). Metode DSR digunakan peneliti untuk menemukan solusi atas permasalahan-permasalahan aktual. Pengumpulan data dilakukan dengan studi dokumen, wawancara dan observasi. Fokus area yang diukur antara lain Business, People, Process dan Technology. enelitian ini menghasilkan penilaian beberapa area yang perlu ditingkatkan untuk mengamankan sistem keamanan informasi. Penilaian kembali diperoleh beberapa area telah dilaksanakan sesuai rekomendasi HPE, namun masih terdapat 38 aktivitas yang belum dilaksanakan. Hasil penilaian secara mandiri menggunakan tools SOC-CMM diperoleh pencapaian tingkat kematangan di angka 1.3 (Initial) meningkat dibandingkan dengan hasil pengukuran HPE sebesar 0.78 (Incomplete). Penilaian berdasarkan framework NIST Cybersecurity diperoleh 16 area yang belum tercapai. Dibutuhkan pengembangan SIEM lebih lanjut untuk mengantisipasi secara dini ancaman. Pembangunan fasilitas SOC saat ini diperlukan, namun sebelumnya perlu dilakukan terlebih dahulu peningkatan beberapa area untuk mematangkan sistem keamanan PT XYZ.

---

ABSTRACT nformation Technology Strategic Plan PT XYZ is committed to implementing a system of security and monitoring in the form of developing IT infrastructure security reliability. But in reality there were still security incidents that resulted in financial and non-financial losses, in addition the results of measurements by Hewlet Packard Enterprise (HPE) in 2016 stated that the maturity level of PT XYZs security operations was on incomplete level, so PT XYZ had problems managing security operations not yet feasible. This study evaluates the current management of security operations based on previous HPE measurement results and independently measures maturity level using SOC-CMM tools. The design of the Security Operations Center (SOC) uses the Design Science Research (DSR) approach. The DSR method is used by researchers to find solutions to actual problems. Data collection is done by document study, interview and observation. The focus of the area measured includes Business, People, Process and Technology. This research resulted in the assessment of several areas that need to be improved to secure information security systems. Re-assessment obtained by several areas has been carried out according to HPE recommendations, but there are still 38 activities that have not been implemented. The results of the self-assessment using SOC-CMM tools obtained the achievement of the maturity level at 1.3 (Initial) increased compared to the HPE measurement result of 0.78 (Incomplete). Assessments based on the NIST Cybersecurity framework obtained 16 areas that have not been reached. Further development of SIEM is needed to anticipate threats early. The construction of the SOC facility is currently needed, but beforehand it was necessary to do an increase in several areas to finalize the security system of PT XYZ.