[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-127064-en":3,"doc-seo-127064-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},127064,962084931830,"Theodore","https://ap-avatar.wpscdn.com/davatar_a8503ba1806abce46bf441b54a3ca4cd",8,"Research & Report","Machine Learning on Encrypted Data - Analyzing Efficiency and Accuracy Trade-Offs","Modern phishing poses a serious threat by leaking personal information from emails and websites, making stronger detection essential. Machine learning is proposed as a predictive approach using historical data to identify phishing behavior and improve countermeasures. The document analyzes methods for phishing detection, discusses machine-learning algorithms used for classification, and describes data collection and processing. It also evaluates results demonstrating how the approach supports accurate detection while considering efficiency and accuracy trade-offs.","Machine Learning on Encrypted Data: Analyzing Efficiency and  \nAccuracy Trade-Offs  \nKavitha Kumari  \nResearch Scholar, Amity University, Pune, INDIA.  \n[www.jrasb.com || Vol. 2 No. 4](www.jrasb.com || Vol. 2 No. 4) (2023): August Issue  \nReceived: 21-07-2023 Revised: 01-08-2023 Accepted: 18-08-2023  \nABSTRACT  \nIn this modern era, phishing has become a great problem. Because of this, it can be observed that the personal information of people is leaked from emails & websites. Hence, it is needed that these instances of phishing are to be reduced. In doing one of the best tools that can be used is Machine Learning. This is a process of using historical data for making prediction of future scenarios. In this project, the details of the approached that can be used for the detection of phishing are analyzed. Moreover, the algorithms that are used by ML for this purpose are also envisaged here. The description of the process of collection of data is presented here. In addition to this, the results that shows the effectiveness of ML in the detection of phishing is also discussed here.  \nKeywords-Machine Learning, Neural Network, Decision Tree, Random Forest, Feature Engineering.  \nI. INTRODUCTION  \nOne of the most critical threats that are observed these days is the threat of the stealing of information through websites. In this way, the personal info of the users gets stolen. One of the traditional methods for the detection of this problem is the blacklisting of such sites. However, it has become irrelevant as the number of sites has increased a lot over the past few years. The best tool that can be used in this current scenario is “Machine Learning”.  \nThis has the feature of checking a huge amount of data and analyzing them to find out the websites that are pure and have no risks of leaking information. The things that are analyzed in this process are the content present in the email, URLs, & the source codes. Here, the details of the process of “phishing detection” are discussed. For this, literature based on the detection of phishing was studied in order to identify the processes of this.  \nII. LITERATURE REVIEW  \n2.1 Machine Learning-based solutions for phishing website detection  \nAccording to Tang & Mahmoud, 2021, the use of ML shows good results in terms of detecting phishing on websites. It is considered to be one of the most critical threats or the users. There are different models of ML that can be used for this purpose. This contains both“supervised” & “unsupervised” models. The data that these models analyses are collected from the URLs, and content of the websites. “Supervised learning” has attained success in the detection of “phishing websites”. The approaches that it uses are “decision tree”,“neural network”, and SVM. This is based on the datasets with the help of which it is predicted that how often a website can be phishing (Tang & Mahmoud, 2021) . These processes can easily be interpreted and their accuracy is fair enough. In the current era, the use of CNNs & RNNs has become very popular. The results are good in terms of detection of phishing. The main benefit of these is that they are able to collect necessary information from the existing data. This lowers the need for “feature engineering”. The main  \ncharacteristic of CNN is that it can differentiate a good and a phishing website visually. On the other hand, the use of RNN is mainly observed in the processing of“sequential data”.  \n2.2 Phishing URL detection using lexical-based machine learning in a real-time environment  \nAccording to Gupta et al. 2021, the prime focus of this method is on finding out the composition & structure of URLs. In this way, phishing is detected. The process includes checking out the length of the URL, looking for the presence of keywords that are suspicious in nature and also the presence of “special characters”. The “lexical features” are such things that can easily be extracted and also analysed for the detection of phishing in","cbCaio2xJBcqb5Jk","https://ap.wps.com/l/cbCaio2xJBcqb5Jk","pdf",456411,1,13,"English","en",105,"# Abstract\n# Introduction\n# Literature Review\n## Machine Learning-based solutions for phishing website detection\n## Phishing URL detection using lexical-based machine learning in a real-time environment\n# Methods\n## Data collection & processing","[{\"question\":\"Why is phishing detection considered important in the presented approach?\",\"answer\":\"Phishing can leak users’ personal information through emails and websites. The document treats phishing as a critical threat and motivates reducing these incidents using ML-based detection.\"},{\"question\":\"What data sources and processing steps are used before training and evaluation?\",\"answer\":\"Data consists of URLs containing both benign and phishing examples, collected from public phishing databases and crawlers. The data is then pre-processed before analysis to support training and evaluation.\"},{\"question\":\"Which machine learning techniques are discussed for phishing detection?\",\"answer\":\"The document reviews supervised ML methods such as decision trees and neural networks, and mentions approaches using lexical features for URLs and log data patterns. It also references ensemble methods like random forest and feature engineering concepts.\"}]","Machine Learning on Encrypted Data - Analyzing Efficiency and Accuracy Trade-Offs | PDF",1785936620,33,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"machine-learning-on-encrypted-data-analyzing-efficiency-and-accuracy-trade-offs","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/machine-learning-on-encrypted-data-analyzing-efficiency-and-accuracy-trade-offs/127064/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-05",true,{"@type":65,"interactionType":66,"userInteractionCount":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"Why is phishing detection considered important in the presented approach?","Question",{"text":75,"@type":76},"Phishing can leak users’ personal information through emails and websites. The document treats phishing as a critical threat and motivates reducing these incidents using ML-based detection.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What data sources and processing steps are used before training and evaluation?",{"text":80,"@type":76},"Data consists of URLs containing both benign and phishing examples, collected from public phishing databases and crawlers. The data is then pre-processed before analysis to support training and evaluation.",{"name":82,"@type":73,"acceptedAnswer":83},"Which machine learning techniques are discussed for phishing detection?",{"text":84,"@type":76},"The document reviews supervised ML methods such as decision trees and neural networks, and mentions approaches using lexical features for URLs and log data patterns. It also references ensemble methods like random forest and feature engineering concepts.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]