Personalized Search Engine Research Articles

Document Difficulty Aspects for Medical Practitioners: Enhancing Information Retrieval in Personalized Search Engines

Timely and relevant information enables clinicians to make informed decisions about patient care outcomes. However, discovering related and understandable information from the vast medical literature is challenging. To address this problem, we aim to enable the development of search engines that meet the needs of medical practitioners by incorporating text difficulty features. We collected a dataset of 209 scientific research abstracts from different medical fields, available in both English and German. To determine the difficulty aspects of readability and technical level of each abstract, 216 medical experts annotated the dataset. We used a pre-trained BERT model, fine-tuned to our dataset, to develop a regression model predicting those difficulty features of abstracts. To highlight the strength of this approach, the model was compared to readability formulas currently in use. Analysis of the dataset revealed that German abstracts are more technically complex and less readable than their English counterparts. Our baseline model showed greater efficacy than current readability formulas in predicting domain-specific readability aspects. Conclusion: Incorporating these text difficulty aspects into the search engine will provide healthcare professionals with reliable and efficient information retrieval tools. Additionally, the dataset can serve as a starting point for future research.

Xperia Search Engine

Abstract: This research paper delves into the innerworkings of search engines and introduces Xperia, a personalized search engine aimed at enhancing information retrieval. The paper explores the fundamentaltechnologies employed by search engines, tracing their evolution and growth over time. It presents the development and implementation of Xperia, highlighting its unique feature set that goes beyond traditional search engines by providing users with not only relevant resourcelinks but also extracted information from various web sources. The paper begins with an introduction, providing the background and motivation for the research, as well as outlining the research objectives and scope. It then delves into the fundamentals of search engines, discussing their components, crawling and indexing processes, ranking algorithms, and user interfaces. The evolution of search engine technologies is examined, from the early stages to the current advancements in semantic search, natural language processing, and the incorporation of machine learning and artificial intelligence.

User Modeling and Profiling in Information Systems

User modeling or user profiling is fundamental to manage information overload issues in many adaptive and personalized systems (e.g., recommender systems, personalized search engines, adaptive user interfaces). Although there are some literature review papers that provide an overview of existing studies in user modeling and their usage, there is currently a lack of bibliometric studies that can provide a systematic and quantitative overview of this research area. Therefore, this paper aims to complete the existing literature in this research area through a bibliometric study based on 52,027 relevant publications extracted from Scopus, a world-leading publisher-independent global citation database. The analyses enabled us to identify the most relevant publications, sources of publications, authors, institutions, countries, and their collaboration. We also identify and classify the twelve most important associated topics, along with their subtopics and their trends. Some identified weak signals in topic trend analysis also provide good ideas of potential future research directions.

The Answer Bot Effect (ABE): A powerful new form of influence made possible by intelligent personal assistants and search engines.

We introduce and quantify a relatively new form of influence: the Answer Bot Effect (ABE). In a 2015 report in PNAS, researchers demonstrated the power that biased search results have to shift opinions and voting preferences without people’s knowledge–by up to 80% in some demographic groups. They labeled this phenomenon the Search Engine Manipulation Effect (SEME), speculating that its power derives from the high level of trust people have in algorithmically-generated content. We now describe three experiments with a total of 1,736 US participants conducted to determine to what extent giving users “the answer”–either via an answer box at the top of a page of search results or via a vocal reply to a question posed to an intelligent personal assistant (IPA)–might also impact opinions and votes. Participants were first given basic information about two candidates running for prime minister of Australia (this, in order to assure that participants were “undecided”), then asked questions about their voting preferences, then given answers to questions they posed about the candidates–either with answer boxes or with vocal answers on an Alexa simulator–and then asked again about their voting preferences. The experiments were controlled, randomized, double-blind, and counterbalanced. Experiments 1 and 2 demonstrated that answer boxes can shift voting preferences by as much as 38.6% and that the appearance of an answer box can reduce search times and clicks on search results. Experiment 3 demonstrated that even a single question-and-answer interaction on an IPA can shift voting preferences by more than 40%. Multiple questions posed to an IPA leading to answers that all have the same bias can shift voting preferences by more than 65%. Simple masking procedures still produced large opinion shifts while reducing awareness of bias to close to zero. ABE poses a serious threat to both democracy and human autonomy because (a) it produces large shifts in opinions and voting preferences with little or no user awareness, (b) it is an ephemeral form of influence that leaves no paper trail, and (c) worldwide, it is controlled almost exclusively by just four American tech companies. ABE will become a greater threat as people increasingly rely on IPAs for answers.

IP Geolocation through Geographic Clicks

IP geolocation databases map IP addresses to their physical locations. They are used to determine the location of online users when their precise location is unavailable. These databases are vital for a number of online services, including search engine personalization, content delivery, local ads, and fraud detection. However, IP geolocation databases are often inaccurate. In this work we present two novel approaches to improving IP geolocation by mining search engine click logs. First, we show that we can derive which URLs have local affinity by clustering clicks from IPs with known locations. We demonstrate that we can further propagate these URL locations to IP addresses with unknown locations. Our approach significantly outperforms two state-of-the-art commercial IP geolocation databases by 25 and 36 percentage points at a distance error of 10 kilometers, respectively. Second, we present an alternative method of assigning locations to URLs when IP location training data is not available, by instead extracting locations from the body of web documents. This second approach also outperforms the baselines by 7 and 17 percentage points, respectively, and has higher coverage than the first method. Finally, we also demonstrate that our two approaches outperform the academic state of the art based on mining query logs.

Curating Viking objects through customized scrolling: How search engines personalize historical narratives

The stories that museum objects convey change on social media. This narrative shift is analyzed through the development of social media methods that track the dissemination of the Viking helmet from the Swedish History Museum to personalized YouTube search engine result pages (SERPs). Social media methods are used to compare the narratives and meanings produced on YouTube SERPs with the narratives and meanings sanctioned by the Swedish History Museum. The hermeneutic process at the core of the methodological approach reveals how personalization algorithms targeting user identities through IP addresses influence the meanings of objects, which are mediated and commercialized via social media SERPs. This research demonstrates that personalized search engine result pages have an algorithmic impact on the narratives of objects that counters the social justice goals integral to contemporary museum practices. The meaning curation of museum objects is not customized to individual preferences and identities. Museums advance selected interpretations of objects to endorse a variety of vantage points so that the fixity of identity categories can be questioned.

Büyük Veri, Yapay Zeka ve Makina Öğrenimi Algoritmalar: Hakikat Ötesi Çağda Dijital Tehditlerin Betimleyici bir Analizi

Did the utilization of big data change everything about how information circulates? Our digital data have been kept in big warehouses that we name ‘big data’. All the things that we do in virtual life leave a digital footprint and thanks to machine learning algorithms; in our newsfeed, we mostly face contents, which are similar to the subjects that we looked for before. Big data are being used for manipulating people to buy new products, to travel to new places, to read new books, etc. However, as it emerged in 2016 with Cambridge Analytica Scandal of Facebook, sometimes those technologies construct a threat for democracy. The underlying reason is that in our days, big data and AI algorithms have been used by political campaign managers to manipulate and/or persuade people through diffusion of promoted ‘false’ contents. The aim of this study, by doing a descriptive analysis of very recent historical events like the failure of Microsoft’s AI Tay and Youtube’s effects on the presidential election in Brazil, is to define very current digital threats against democracy. Additionally, to better describe and discuss these digital threats we did semi-structured interviews with four experts who work on AI algorithms, big data, and social engineering. Our analyses and findings that we gathered from semi-structured interviews showed that there are several digital threats in the post-truth era that we live in like digital manipulation, violation of data privacy and misuse of big data, personalized search engine algorithms that create filter bubbles, the ease of production and diffusion of fake contents.

Collaborative Search Engine for Enhancing Personalized User Search Based on Domain Knowledge.

In the fast moving world, users cross over large amount of data for their daily life. Due to the misinterpretation of the context, user cannot retrieve the proper context or failure to retrieve the information. The main aim of this paper is to design and implement a personalized search engine which works based on the domain of the user with the specific constraints suggested by the user. In this paper, the proposed system, build a search engine with web content which get information from the document corpus for the domain through the cloud databases. Web search engine re-ranks the generic results based on a ranking of a context linked with the domain. In this system, collaborative search service helps to improve the relevancy of the search results and to reduce the overtime on bad links and hence caters to customized needs with collaborative feedback using fuzzy decision tree based on fuzzy rules.

A comprehensive mechanism for hotel recommendation to achieve personalized search engine

A comprehensive mechanism for hotel recommendation to achieve personalized search engine

Improving statistical relational learning with graph embeddings for socio-economic data retrieval

Social media data is useful for personalized search engines, recommender systems, and targeted online marketing. Sometimes values of attributes are missing due to security reasons or problematic data collection process. In this case, the information about connections between vertices become more important since it explicitly allows for using the structure of a social graph for inferring missing attributes. One of the general and effective approaches of inferring missing attributes on graph structures are statistical relational learning. For machine learning tasks, the graph embeddings represent topological properties better, but they are not aimed at the attributes prediction. In this study, we introduce a method combining graph embeddings and the statistical relational learning. We consider different their combinations, as there are possible different hidden connections between social ties and considered attributes. We compare the performance using real data from the social network with different missing attributes and assortative patterns: gender, age, and economic status. As a result, the inclusion of graph embeddings in statistical relational learning improves accuracy and significantly decreases the number of iterations.

A comprehensive mechanism for hotel recommendation to achieve personalized search engine

A comprehensive mechanism for hotel recommendation to achieve personalized search engine

Trapped in the Filter Bubble? Exploring the Influence of Google Search on the Creative Process

This article explores the use of Google search by art directors and copywriters when developing creative advertisements. It aims to reveal the influence of the widely applied search engine on creative process actions, including information gathering and ideation. In-depth interviews with practitioners are analyzed with reference to literature on advertising creativity and search engine personalization. This exploratory study argues that while serving as a convenient means of accessing digital media content, the personalization capabilities of Google search limit exposure to the qualitatively diverse information needed to trigger new ideas. As such, this article provides valuable insight into the complexities of producing advertisements in the digital media era and highlights avenues for further research into the information-gathering strategies of creative practitioners.

A study on the personalization methods of the web

Search engine personalization is one of the various deep personalization methods. It can be said that personalization systems that help users find the information they need requires the use of contextual and semantic information analysis techniques that exist in the field of data recovery such as web personalization and the process of optimizing the methods to get to web pages in a way that are consistent with the needs of each user. What helps the current problem of search engines and accelerate their performance, is providing a proper framework for finding the correct pattern considering great items in history of users. This approach improves the advising process of the search engines as well. The aim of this paper is to introduce some process improvement methods of correct patterns and analyze them. Here we will discuss the basic concepts of web personalization and consider the three approaches of web personalization and we evaluated the methods belonging to each of them. Keywords: personalization, search engine, user preferences, data mining methods

English

English

Personalized Search Engine for Mobiles Using Location Accuracy and Privacy

Personalized Search Engine for Mobiles Using Location Accuracy and Privacy

Technologies for Streaming Processing of News for Personalized Search Engine Results for News Content

This article describes the implementation of an approach to personalized delivery of the news that considers the content and other characteristics of news reports, prominence, relevance of news events, and the user's personal preferences. It also presents the results of the personalized delivery in a structured, dynamic, and convenient way.

Personalized Search of User Search Behaviourwith Ontology

In this paper, we introduce “Ontology and generic programming” to understand user search behaviors. Personalized search is an important research area that aims to resolve the ambiguity of query terms. To increase the relevance of search results, personalized search engines create user profiles to capture the users’ personal preferences and as such identify the actual goal of the input query. Since users are usually reluctant to explicitly provide their preferences due to the extra manual effort involved, recent research has focused on the automatic learning of user preferences from users’ search histories or browsed documents and the development of personalized systems based on the learned user preferences. Most personalization methods focused on the creation of one single profile for a user and applied the same profile to all of the user’s queries. We believe that different queries from a user should be handled differently because a user’s preferences may vary across queries. In this paper, we conduct extensive analyses and comparisons to evaluate the effectiveness of ontology in several search applications: determining user satisfaction, predicting user search interests, and suggesting related queries. Experiments on large scale datasets of a commercial search engine show that: (1) ontology performs better than session, query and task trails in determining user satisfaction; (2) Ontology increases web page utilities of end users comparing to session, query and task trails ; (3) generic programming is more sensitive than other trail methods in measuring different ranking functions; (4) Query suggestion based on ontology is a good complement of query suggestions based on session trail and click-through bipartite. The findings in this paper verify the need of extracting ontology from web search logs and enhance applications in search and recommendation systems.

Quest Trail: An Effective Approach for Construction of Personalized Search Engine

Personalized search refers to search experiences that are tailored specifically to an individual's interests by incorporating information about the individual beyond specific query provided. Especially people working in a software development organization (analysts, developers, testers, maintenance team members), find it increasingly difficult to get relevant results to their searches. We propose methods to personalize searches by resolving the ambiguity of query terms, and increase the relevance of search results in order to match the user‟s interests. Difficulty in web searches has given rise to the need for development of personalized search engines. Personalized search engines create user profiles to capture the users‟ personal preferences and as such identify the actual goal of the input query. Since users are usually reluctant to explicitly provide their preferences due to the extra manual effort involved, the search engine faces the entire burden of predicting the user‟s preferences and intentions behind a query in order to yield more relevant search results. In this paper we define a QUEST to be the objective of user‟s search; here we combine quest level analysis of user‟s search logs and semantic analysis of the user‟s query in order to personalize user‟s search results. Most personalization methods focus on the creation of one single profile for a user and apply the same profile to all of the user‟s queries. Hence we propose a personalized search for a software development organization by creating QUEST or domain based profile rather than individual user based profile.

Personalized Search Engine using Social Networking Activity

The main objective of this research work is to obtain a personalized search result required by the user, by creating user profile based on social networking activity. The user profile is actually constructed by pages liked by the individual user in Facebook on their respective user account. According to the constructed user profile the results are re-ranked and the personalised search results are obtained. Lingo- a novel algorithm is used for clustering the data. The search results are retrieved to user using Carrot2 API search engine. In the past, personalized search engines acquired data from surfing history implicitly or explicitly by machine learning whereas this work acquires data implicitly through user likes and also explicitly through user defined categories . The Facebook likes are given by each user only on their personal interest. Thus, these data play a vital in providing accurate search results to each user and provide exact search results as per the user interest. Keywords: Information Retrieval, Personalization, Search Engine, Social Network and Web Search

PSEFiL: A Personalized Search Engine with Filtered Links

dynamic nature of the World Wide Web and its growing dimension make retrieving the exact information a difficult task. Inaccurate answers delivered by search engines especially for the query phrases with different meaning makes the feeling of dissatisfaction in today's surfers who needs the specific answer for their information demand. Search engines nowadays tries to understand users' request through studying his/her search background or even make users participate in the search process in order to clarify what he/she really needs. This trend is part of the search engines' endeavors to become personalized. One of the well-formed personalized search engines is SNAKET that employs the user participation for personalization process. In this paper based on the personalization algorithm of SNAKET, we propose an architecture of our personalized search engine of PSEFiL that filters out the links and delivers the answers to users with low or no content drift as a means of enriching the answer set. Furthermore, the answer set is robust because every existing link in result set either is highly ranked from other search engines or has no or less subject jumps with an exact manual scan process. Also every link is clearly classified to every fetched meaning of a query phrase. One objective of PSEFiL is to give the accurate answers not to populate the answer set with more links that may contain less or no accurate answers. Keywordsengine, Search engine optimization, Search engine personalization, Web Structure Mining.