Improving Recommendation Quality in Google Drive

Abstract

Quick Access is a machine-learned system in Google Drive that predicts which files a user wants to open. Adding Quick Access recommendations to the Drive homepage cut the amount of time that users spend locating their files in half. Aggregated over thẽ 1 billion users of Drive, the time saved up adds up to~1000 work weeks every day. In this paper, we discuss both the challenges of iteratively improving the quality of a personal recommendation system as well as the variety of approaches that we took in order to improve this feature. We explored different deep network architectures, novel modeling techniques, additional data sources, and the effects of latency and biases in the UX. We share both pitfalls as well as successes in our attempts to improve this product, and also discuss how we scaled and managed the complexity of the system. We believe that these insights will be especially useful to those who are working with private corpora as well as those who are building a large-scale production recommendation system. CCS CONCEPTS • Information systems → Recommender systems; • Computing methodologies → Neural networks.

Peningkatan Mutu Pelayanan Sistem Kinerja Dosen di Kopertais Wilayah IV Surabaya

“…Quality Function Deployment (QFD) merupakan metode perencanaan dan pengembangan produk yang memiliki tujuan untuk menerjemahkan kebutuhan pelanggan (Ocampo dkk., 2020) dan memenuhi kebutuhan pelanggan sesuai dengan sumber daya dan kemampuan yang dimiliki saat ini (Yazdani dkk., 2019), maka dari itu QFD dianggap sebagai metode yang sangat berguna untuk meningkatan produk, perencanaan kualitas, dan sebagai dasar pengambilan keputusan (Pandey, 2020), QFD sendiri berasal dari jepang pada tahun 1972 dan digunakan oleh perusahaan seperti Toyota dan Mitsubishi untuk memperbaiki kualitas produk yang dihasilkan. (Akkawuttiwanich dkk., 2018) QFD biasa dilakukan dengan membuat matriks korelasi antara kebutuhan dari pelanggan dan desain jasa atau produk berupa HOQ (house of quality) (Chen, 2020). HOQ menyampaikan informasi yang cukup detail mengenai apa yang diharapkan pelanggan serta bagaimana persyaratan dari pelanggan mampu untuk diidentifikasi.…”

ORCA: A Network and Architecture Co-design for Offloading us-scale Datacenter Applications

“…DLRMs have received much attention by Internet giants [2,24,36,57,58,82,117] as they can offer more revenue and better user experience. In an end-to-end recommender system, the most expensive part of serving an inference request is the embedding reduction step, consuming huge memory capacity [117,177,178] and 1 2 ∼ 3 4 of the inference time [38,58,60,82,92].…”

Securing Academic Student File Using Aes Algorithm For Cloud Storage Web-Based System

“…The development of the system aims to assist users in backup and to save their files in the cloud in secure ways. Recently, users preferred to backup and save files in cloud storage such as Google Drive, iCloud, and Dropbox (Chen et al, 2020). However, the current application has been exposed to certain unauthorised access and misused risk from external threats (Chen et al, 2020).…”

“…Recently, users preferred to backup and save files in cloud storage such as Google Drive, iCloud, and Dropbox (Chen et al, 2020). However, the current application has been exposed to certain unauthorised access and misused risk from external threats (Chen et al, 2020). This occurred when private files in the cloud storage were accessed by unauthorised users by mistake or intentionally after successfully breaching the entry password (Goel & Hofstede, 2021).…”