CYCLOSA: Decentralizing Private Web Search through SGX-Based Browser Extensions

Googling for Health Information

The code base of system software is growing fast, which results in a large number of vulnerabilities: for example, 296 CVEs have been found in Xen hypervisor and 2195 CVEs in Linux kernel. To reduce the reliance on the trust of system software, many researchers try to provide trusted execution environments (TEEs), which can be categorized into two types: non-privileged TEEs and privileged TEEs. Non-privileged TEEs (e.g., Intel SGX) are extensible, but cannot protect security services like virtual machine introspection (VMI) due to the lack of system-level semantics. On the contrary, privileged TEEs (e.g., the secure world of ARM TrustZone) have system-level semantics, but any additional service implemented in the privileged TEE directly increases the TCB of the entire system. In this article, we propose a new design of TEE to support system-level security services and achieve better extensibility with a small TCB. Each TEE instance of the proposed design is named a <small>Colony</small>. Specifically, we introduce a <i>secure monitor</i> for isolation and capability management. Each <small>Colony</small> is assigned capabilities to access only necessary system-level semantics. We use the new TEE to build four security services, including secure device accessing, VMI tools, a system call tracer, and a much more complex service to virtualize ARM TrustZone with multiple <small>Colonies</small>. We have implemented the system on ARMv7 and ARMv8 platforms, in Xen hypervisor and Linux kernel, and perform a detailed evaluation to show its efficiency.<xref rid="fn1" ref-type="fn">¹</xref><fn id="fn1"><label>1.</label> This paper is an extended version of the conference paper published in USENIX Security&#x2019;17: vTZ: Virtualizing ARM TrustZone <xref ref-type="bibr" rid="ref29">[29]</xref> . A brief summary of differences is in Section<xref ref-type="sec" rid="sec8">8</xref> . </fn>

Purpose — The purpose of this chapter is to give an overview of the context of Web search and search engine related research, as well as to introduce the reader to the sections and chapters of the book. Methodology/approach — We review literature dealing with various aspects of search engines, with special emphasis on emerging areas of Web searching, search engine evaluation going beyond traditional methods and new perspectives on Web searching. Findings — The approaches to studying Web search engines are manifold. Given the importance of Web search engines for knowledge acquisition, research from different perspectives needs to be integrated into a more cohesive perspective. Research limitations/implications — The chapter suggests a basis for research in the field and also introduces further research directions. Originality/value of paper — The chapter gives a concise overview of the topics dealt within the book and also shows directions for researchers interested in Web search engines.

Outsourcing Machine Learning (ML) tasks to cloud servers is a cost-effective solution when dealing with distributed data. However, outsourcing these tasks to cloud servers could lead to data breaches. Secure computing methods, such as Homomorphic Encryption (HE) and Trusted Execution Environments (TEE), have been used to protect outsourced data. Nevertheless, HE remains inefficient in processing complicated functions (e.g., non-linear functions) and TEE (e.g., Intel SGX) is not ideal for directly processing ML tasks due to side-channel attacks and parallel-unfriendly computation.In this paper, we propose a hybrid framework integrating SGX and HE, called HT2ML, to protect user's data and models. In HT2ML, HE-friendly functions are protected with HE and performed outside the enclave, while the remaining operations are performed inside the enclave obliviously. HT2ML leverages optimised HE matrix multiplications to accelerate HE computations outside the enclave while using oblivious blocks inside the enclave to prevent access-pattern-based attacks. We evaluate HT2ML using Linear Regression (LR) training and Convolutional Neural Network (CNN) inference as two instantiations. The performance results show that HT2ML is up to ∼11× faster than HE only baseline with 6-dimensional data in LR training. For CNN inference, HT2ML is ∼196× faster than the most recent approach (Xiao et al., ICDCS'21).

Today’s conventional search engines hardly do provide the essential content relevant to the user’s search query. This is because the context and semantics of the request made by the user is not analyzed to the full extent. So here the need for a semantic web search arises. SWS is upcoming in the area of web search which combines Natural Language Processing and Artificial Intelligence. The objective of the work done here is to design, develop and implement a semantic search engine - SIEU(Semantic Information Extraction in University Domain) confined to the university domain. SIEU uses ontology as a knowledge base for the information retrieval process. It is not just a mere keywordsearch. It is one layer above what Google or any other search engines retrieve by analyzing just the keywords. Here the query is analyzed both syntactically and semantically. The developed system retrieves the web results more relevant to the user query through keyword expansion. The results obtained here will be accurate enough to satisfy the request made by the user. The level of accuracy will be enhanced since the query is analyzed semantically. The system will be of great use to the developers and rese archers who work on web. The Google results are re-ranked and optimized for providing the relevant links. For ranking an algorithm has been applied which fetches more apt results for the user query.

Securing sensitive user information and maintaining privacy during authentication and authorization processes is crucial in today’s interconnected digital world. This study investigates new approaches to tackle these obstacles by combining blockchain technology with Trusted Execution Environment (TEE). The proposed system introduces a new method that combines the advantages of both technologies to create a strong and privacy-focused framework for network security. The study assesses the effectiveness of the suggested system based on key parameters, focusing specifically on Transaction Throughput (TPS) and Latency. The combination of blockchain and TEE aims to improve privacy protection and security protocols, providing a dependable solution for privacy-focused authentication and authorization in network settings. The experiments confirmed the efficacy of the proposed model. The Transaction Throughput (TPS) reached 296 transactions per second, demonstrating the system’s scalability and capacity to manage a large number of transactions. The Latency was measured at 54 milliseconds, indicating a fast and responsive authentication and authorization process. The research enhances the development of privacy-preserving solutions by emphasizing the importance of the blockchain and TEE collaboration. The proposed model surpasses performance expectations set by traditional methods, positioning itself as a significant advancement in network security. The results highlight the system’s ability to significantly improve privacy protection in digital transactions and set the stage for advancements in secure, efficient, and privacy-focused authentication and authorization systems. between the quantity demand and the price.

Web search engines have long served as indispensable tools for information retrieval; user behavior and query formulation strategies have been well studied. The introduction of search engines powered by large language models (LLMs) suggested more conversational search and new types of query strategies. In this paper, we compare traditional and LLM-based search for the task of image geolocation, i.e., determining the location where an image was captured. Our work examines user interactions, with a particular focus on query formulation strategies. In our study, 60 participants were assigned either traditional or LLM-based search engines as assistants for geolocation. Participants using traditional search more accurately predicted the location of the image compared to those using the LLM-based search. Distinct strategies emerged between users depending on the type of assistant. Participants using the LLM-based search issued longer, more natural language queries, but had shorter search sessions. When reformulating their search queries, traditional search participants tended to add more terms to their initial queries, whereas participants using the LLM-based search consistently rephrased their initial queries.

Pontis: A decentralized framework for unifying remote attestation and enabling interoperability between heterogeneous TEEs

PurposeThe purpose of this paper is to analyze the relationship between new product diffusion and consumer internet search patterns using big data and to investigate whether such data can be used in forecasting new product diffusion.Design/methodology/approachThis research proposes a new product diffusion model based on the Bass diffusion model by incorporating consumer internet search behavior. Actual data from search engine queries and new vehicle sales for each vehicle class and region are used to estimate the proposed model. Statistical analyses are used to interpret the estimated results, and the prediction performance of the proposed method is compared with other methods to validate the usefulness of data for internet search engine queries in forecasting new product diffusion.FindingsThe estimated coefficients of the proposed model provide a clear interpretation of the relationship between new product diffusion and internet search volume. In 83.62 percent of 218 cases, analyzing the internet search pattern data are significant to explain new product diffusion and that internet search volume helps to predict new product diffusion. Therefore, marketing that seeks to increase internet search volume could positively affect vehicle sales. In addition, the demand forecasting performance of the proposed diffusion model is superior to those of other models for both long-term and short-term predictions.Research limitations/implicationsAs search queries have only been available since 2004, comparisons with data from earlier years are not possible. The proposed model can be extended using other big data from additional sources.Originality/valueThis research directly demonstrates the relationship between new product diffusion and consumer internet search pattern and investigates whether internet search queries can be used to forecast new product diffusion by product type and region. Based on the estimated results, increasing internet search volume could positively affect vehicle sales across product types and regions. Because the proposed model had the best prediction power compared with the other considered models for all cases with large margins, it can be successfully utilized in forecasting demand for new products.

Geography is becoming increasingly important in web search. Search engines can often return better results to users by analyzing features such as user location or geographic terms in web pages and user queries. This is also of great commercial value as it enables location specific advertising and improved search for local businesses. As a result, major search companies have invested significant resources into geographic search technologies, also often called local search.This paper studies geographic search queries, i.e., text queries such as hotel new york that employ geographical terms in an attempt to restrict results to a particular region or location. Our main motivation is to identify opportunities for improving geographical search and related technologies, and we perform an analysis of 36 million queries of the recently released AOL query trace. First, we identify typical properties of geographic search (geo) queries based on a manual examination of several thousand queries. Based on these observations, we build a classifier that separates the trace into geo and non-geo queries. We then investigate the properties of geo queries in more detail, and relate them to web sites and users associated with such queries. We also propose a new taxonomy for geographic search queries.

This article tries to analyse, and measure the uncertainty associated with the relevant document selection in web-search engines. The number of index terms, and their occurrence frequency influences the relevance calculation. The variation in term frequency either in processed web documents or in users' query influences the relevance calculation. This leads to an uncertainty associated with the document selection, and it is relevance calculation. In this article, we proposed a new measure called entropy. The entropy can be measured by varying the documents' term frequency or user's query term frequency. As the web documents can't be changed, we used variation in user's query term frequency to measure the uncertainty associated with the document selection in web-search engines. The query's term frequency is varied and given to the search engines. namely 'Google', and 'Bing' The high uncertainty gives scope for search engine optimisation. From the high uncertainty search engines, we can extract more relevant documents.

Internet has become an instant source of informatio n for almost anyone. By viewing the results of sear ch terms displayed by an Internet Search Engine (ISE), a person may decide his next course of action whether to continue using the Internet, abort or to combine it with other sources of information. The results produced by an ISE can be regarded as an in dex of relative availability of references. Given a list of suitable search terms, a user will firstly have to decide on the ISE to be used. Due to huge potent ial references available for given search terms the use r has to create heuristics to choose the entries sh own on the computer screen. As there are varying breadths and depths of information revealed by various Inter net Search Engines (ISE’s), this paper will not attempt to make a comparison among ISE’s, rather will focu s only on a particular search engine and ascertain th e results it produces given a list of search terms. Google is chosen as a proxy, being one of the most popular ISE’s. By confining to only a search engine, the s tudy affords to control variability among ISE’s should m ultiple ISE’s be used. With the search engine place d under control, it is easy to achieve the primary ob jective of the study, i.e., to ascertain availabili ty of relative breadth of sub-themes of Analytic Hierarch y Process (AHP) in Google. It is natural for user, especially researcher to be concerned with number, quantity. If there is seemingly abundant literature , one would be motivated to pursue research along the the me or sub-theme. In this study, the strength of presence of a sub-theme is measured by using two measures: (i) result of a sub-theme of AHP over the sub-theme itself, and (ii) result of the sub-theme over the total results generated for all of the thi rty six sub-themes used in the search. This study controlle d biasness in specifying the sub-themes of AHP by adopting the sub-themes or search terms specified b y the 2013 AHP conference organizers. This decision helps make the study efficient without with it has to distill the sub-themes by surveying the AHP literature. The data for analysis was gathered by s urfing Google on 26 Feb 2013 8.55 p.m. - 9.26 p.m. Peninsular Malaysian time. The search results were computed to generate two types of ratios specified earlier. A composite index was created using the re sulting two types of ratios which are used to class ify the efficiency, hence dominance of the original res ults (hits). Kendall’s correlation produced statist ically significant correlations between the composite inde x and Rank of AHP specific and area results. Using indices greater than 1.000 as the base, 6 AHP specific areas occupy the top positions with ratios ranging from 19.341 to 61.574; 15 AHP specific areas occupy the second top positions with ratios ranging from 1.119 to 9.602, and 14 AHP specific areas occupy the third and last position with indi ces below 1.000 ranging from 0.050 to 0.894. The paper includes dis cussion, implications, limitations, conclusions and suggestions for further research.

Nowadays, as the usage of Internet has incredibly increased, web search engines become the common approach to find and retrieve needed information. Hence, evaluating search engine quality is a hot topic which attracts many researches' attention. In this paper, we propose a framework named Similarity based Automatic Web Search Engine Evaluation, SAWSEE, to evaluate web search engines. SAWSEE measures the information retrieval effectiveness of web search engines by comparing and voting their returned results, particularly using nDCG metric to rank search engines. SAWSEE compares the search engines' results based on their similarity which is calculated in two consecutive levels, the web page address level and the main content of web page level. To find the similarity of the main content of two search engines' results, SAWSEE utilizes Winnowing algorithm, a well-known and widely-used plagiarism detection method. We compared our method with the results acquired from human assessors' evaluations. The promising comparison shows that SAWSEE provides rankings that are consistent with the rankings resulted from human assessors' evaluations. Hence, the proposed method can be applied in real world environments for evaluation of web search engines.

Search is the most heavily used web application in the world and is still growing at an extraordinary rate. Understanding the behaviors of web search engines, therefore, is becoming increasingly important to the design and deployment of data center systems hosting search engines. In this paper, we study three search query traces collected from real world web search engines in three different search service providers. The first part of our study is to uncover the patterns hidden in the query traces by analyzing the variations, frequencies, and locality of query requests. Our analysis reveals that, contradicted to some previous studies, real-world query traces do not follow well-defined probability models, such as Poisson distribution and log-normal distribution. The second part of our study is to deploy the real query traces and three synthetic traces generated using probability models proposed by other researchers on a Nutch based search engine. The measured performance data from the deployments further confirm that synthetic traces do not accurately reflect the real traces. We develop an evaluation tool that can collect performance metrics on-line with negligible overhead. The performance metrics include average response time, CPU utilization, Disk accesses, and cycles-per-instructions, etc. The third of our study is to compare the search engine with representative benchmarks, namely Gridmix, SPECweb2005, TPC-C, SPECCPU2006, and HPCC, with respect to basic architecture-level characteristics and performance metrics, such as instruction mix, processor pipeline stall breakdown, memory access latency, and disk accesses. The experimental results show that web search engines have a high percentage of load/store instructions, but have good cache/memory performance. We hope those results presented in this paper will enable system designers to gain insights on optimizing systems hosting search engines.

When we search information on the Web, we usually enter a query that best represents our information need, and then modify it as the search results are shown. Many thus have studied search queries to understand how users express and find their information need on the Web. For example, studies have shown that query length affects search performance (Belkin et al, 2003). Then, researchers have begun to examine image search queries on the Web. They analyzed search logs from general or commercial search engines or studied specific image database systems based on log analysis (Goodrum & Sprink, 2001; Jorgensen & Jorgensen, 2005; Goodrum, Bejune, & Siochi, 2003; Pu, 2005; Cunningham, Bainbridge, & Masoodian, 2004). While these studies provide basic understanding of image query formulation, further characteristics of user's queries in the interactive searching process have not been explored. This study thus builds on previous studies on image search queries by examining the patterns of query formulation and modification in visual information search. Specifically, first, it investigates the effects of tasks and other contextual factors on query length and query changes; and second, it compares the query length and changes with the user's satisfaction of search results. Participants Twenty-nine college students were recruited for data collection during the spring and summer semesters of 2008, mostly from the Department of Media Studies at the Catholic University of America (CUA). They included 22 from Media Studies and 7 from other fields taking courses in the department; 22 were females and 7 males; and an average age of 21 years. Background information collected were years of computer experience, hours per day using the Web, self-rated online searching expertise, and self-rated digital image searching experience. The seven-point Likert scale was used for self-rating variables. Each participant took part in three different search sessions with his or her own search topic at CUA with the researcher's laptop that had an Internet connection. The university's homepage was set as the default home for the Web browser. Each participant filled out a pre-search questionnaire to collect a search task, topic, self-rated topic familiarity using 7-Likert scale. The participants chose their own preferred browser between Internet Explorer and Firefox. The search session was recorded by Camtasia v5.0 as screen capture. After the session, a post-search questionnaire was given to the participants to assess the level of satisfaction and confidence of their searching using 7-Likert scale. There was no time limit to complete a search session. The participants were free to end a search session at any time. Search Tasks The participants brought their own task and topics to find images. There were three types of search tasks identified: academic task, work-related task, and personal interest (Table 1). Search queries were acquired from screen capture log files. A total of 979 queries were obtained as the entire string of terms submitted by the participants. Table 2 shows a distribution of search queries on a general search engine, a search engine's image search, and other specific sites. Findings showed that Google Images Search was the most frequent source for forming a search query. An average word per query was 3.07 during the first session, 2.95 during the second session, and 2.89 for the last session. This finding of short queries is similar to that of previous studies on Web image queries (Goodrum & Spink, 2001; Jorgensen & Jorgensen, 2005). Analysis of data is in progress. A poster will summarize detail findings and provide answers to the following questions: What are the patterns of query modification? Are there any differences of query length and number of query change based on the task and topic familiarities? Is there any relationship between query length and number of query change and users' satisfaction and confidence with search results? Do different stages of a user's particular task affect query length and number of query change? The author would like to thank OCLC/ALISE for their generous funding for this project.