Web Search Personalization during the US 2020 Election

Local journalism is in decline in the United States. As one way to assess the state of local journalism, this study examines the information flows between local and national media in the online media ecosystem. Focusing on metropolitan journalism, this paper empirically investigates whether and when a city’s local news coverage can influence national news portrayal of the city. This research draws from intermedia agenda setting (IAS) theory and examines a large news data set related to the most populated 21 US cities. The results suggest that local media are not more likely to transfer the salience of their urban issues to the national media agenda than reversely. In addition, a city’s economic power and the scale of its local journalistic infrastructure, especially the traditional media sector, are significantly correlated with its local media’s power to determine how the city is portrayed in the national media.

During the summer of 2016, the largest cruise ship ever to transit the Northwest Passage traveled through the Canadian Arctic, carrying more than 1,600 passengers and crew. That event resulted in a tremendous influx of tourists, considering the population the entire territory of Nunavut is only 36,000. This study examines how local media played a critical role in representing local community values in coverage of that remarkable cruise. The complete English Canadian textual coverage of the voyage by three media segments was compiled: local media, national media, and a national public broadcaster. A framing analysis revealed significant differences in how each of the three media segments frames the news. Local media placed significantly more emphasis on political, bureaucratic, and environmental frames than did national media, yet used human interest frames at about the same frequency. Furthermore, local media's selection of frames was more diverse than was evident in coverage from national media or the public broadcaster. In addition, the study reveals that the public broadcaster frames stories independently of local and national media, a finding which serves to emphasize the distinct value of each media segment in forming complete coverage of this story.

Guest editors' introduction: special section on mining and searching the web

Introduction. Most university students depend significantly, sometimes exclusively, on the Google search engine for their academic information needs. User satisfaction leads to users’ deeper engagement with an information system that is shown to improve learning in an educational setting. This paper evaluates students’ satisfaction with results from personalised Web search against non-personalised Web search. Method. During semi-structured study sessions, twenty-eight participants (university students) were required to complete a series of search tasks using both personalised and non-personalised Web search. Analysis. Evaluation was based on participants’ explicit feedback as well as their implicit behaviour including search time, number of queries and clicked result links per task, finding the answer and relevance of the search results. Results. There was no apparent significant increase in the participants’ overall level of satisfaction with personalised search results compared to non-personalised results. However, it was found that personalised search reduced the time spent to finish a task and reduced the number of clicks required to arrive at the selected outcome. Conclusions. Personalisation of search results does not increase students' satisfaction with their search results. However, it does reduce the time spent by students in locating information they judged to be satisfactory answers to their questions.

Personalized web search according to user’s geographic and temporal preferences can improve search results quality and satisfy user’s different information needs. We propose a novel approach to capture user’s geographic and temporal preferences in the form of query profile and user preference profile by mining search results and user click-through data leveraging knowledge bases. Our approach classifies queries into five classes based on decision tree algorithm. When personalizing search results, different weights are set to different query classes to balance among content, geographic and temporal information associated with a query. The experiment evaluation results show the effectiveness of our approach and improvement of the search quality.

Most present day search engines have a deterministic behavior in the sense that they return the same search results for all users who submit the same query at a certain time. They do not take the user!s interests and preferences into account in the retrieval process. Integrating user context in the retrieval process can help deliver more targeted search results, thereby providing a personalized search experience to the user. Personalizing web search involves the process of identifying user interests during interaction with the user, and then using that information to deliver results that are more relevant to the user. In this paper, we present our approach to personalizing web search on a mobile device (iPhone). Our approach involves building an ontological model of user interests on the user!s mobile device based on his interaction with web search results. Personalization of search results is achieved by re-ranking search results returned by a standard search engine (Yahoo) based on proximity to the user!s interest model. The ability to recognize user interests in a completely non-invasive way and the accuracy of personalized results are some of the major advantages of our approach.

This study investigates the compliance of Indonesian national (Sindonews.com) and local (Poskota.co.id) online media with the Child Friendly Journalism Guidelines (PPRA) issued by the Indonesian Press Council and UNICEF’s standards on reporting about children. Using quantitative content analysis of 282 child related news articles published in January, June, and December 2024, the research applied 117 indicators grouped into five dimensions: identity protection, language and narrative sensitivity, visual and audio representation, privacy and digital traces, and legal and journalistic ethics. The findings show partial compliance: both outlets performed relatively well in protecting children’s privacy and digital traces but scored lowest in legal and ethical compliance. Identity protection and language sensitivity reflected only moderate adherence, with repeated issues such as disclosure of sensitive details and sensationalist framing across outlets. Statistical analysis confirmed no significant differences between national and local media, suggesting systemic challenges rather than outlet specific weaknesses. These results highlight tensions between commercial pressures and child rights standards in Indonesian journalism. By applying social responsibility theory and framing analysis, this study contributes to scholarly debates on ethical reporting and emphasizes the need for stronger institutional enforcement, participatory monitoring, and trauma informed approaches to safeguard children’s rights in media coverage.

Web search engines are widely used to find huge amount of data on the web in a minimum amount of time. But sometimes it becomes difficult to the users to get exact result for given query. Personalized Web Search (PWS) provides the better search results for individual user needs and improves the quality of the search result based on the user profile. However, user's unwillingness to disclose their private information during search has become major issue in wide increase in PWS. This paper presents a scalable way for users to build rich hierarchical user profiles automatically and provide privacy for the user profiles. Extended User customizable Privacy-preserving Search (E-UPS) framework generalize the user profiles for each query according to user-specified privacy requirements. It will improve the search quality and hide the privacy contents existence in the user profile and gives protection against a typical model of privacy attack.

With a inundating of information in WWW (World Wide Web) users are often failed to retrieve search result in context of their interest through existing search engines. So the personalization of web search result has to be carryout that process user's query and re-rank retrieved results based on their interest. User have diverse background on same query, it is very difficult for some informative query to identify user's current intention. In this paper, a novel approach is proposed that personalize web search result through query reformulation and user profiling. First, a framework is proposed that identify relevant search term for particular user from previous search history by analysing web log file maintained in the server. These terms are appended to user's ambiguous query. Second, the proposed approach proceeds the user's search result and re-rank the retrieved result by identifying interest value of user on retrieved links. Proposed new approach also identify user interest on retrieved links by combing the user interest value generated from VSM (Vector Space Model) and actual rank of that link. Third, the framework also suggest some keywords that help to incorporate user's current interest. Finally, experimental result shows the effectiveness of proposed search engine with commercial search engine with different criteria.

A large part of the modern web is characterized by usergenerated content categorized using collaborative tagging or folksonomy. It becomes difficult to search for relevant content because of ambiguity in lexical representation of concepts and variances in preferences of users. With more and more services relying on tags for content categorization, it is important that search techniques evolve to better suit the scenario. A promising approach towards solving these problems is to use machine common sense in conjunction with folksonomy. A past attempt to use this approach has shown positive results in finding relevant content but it does not address the issue of noise in search results. In this paper, we use the personalized web search technique of traditional web search systems to address the issue of irrelevant search results in common sense and folksonomy based search systems. In personalized web search, results are reflective of user's preferences, which are decided by search history and categories of interest. We propose modifications to personalized web search technique. Using this modified approach, we extend the basic common sense and folksonomy based search systems to address the issue of noise in search results.

It is highly desirable that web search engines know users well and provide just what the user needs. Although great effort has been devoted to achieve this dream, the commonly used web search engines still provide a “one-fit-all” results. One of the barriers is lack of an accurate representation of user search context that supports personalised web search. This article presents a method to represent user search context and incorporate this representation to produce personalised web search results based on Google search results. The key contributions are twofold: a method to build contextual user profiles using their browsing behaviour and the semantic knowledge represented in a domain ontology; and an algorithm to re-rank the original search results using these contextual user profiles. The effectiveness of proposed new techniques were evaluated through comparisons of cases with and without these techniques respectively and a promising result of 35% precision improvement is achieved.

Current conventional search engines deliver similar results to all users for the same query. Because of the variety of user interests and preferences, personalized search engines, based on semantics, hold the promise of providing more efficient information that better reflects users’ needs. The main feature of building a personalized web search is to represent user interests in terms of user profiles. This paper proposes a personalized search approach using an ontology-based user profile. The aim of this approach is to build user profiles based on user browsing behavior and semantic knowledge of specific domain ontology to enhance the quality of the search results. The proposed approach utilizes a re-ranked algorithm to sort the results returned by the search engine to provide a search result that best relates to the user query. This algorithm evaluates the similarity between a user query, the retrieved search results and the ontological concepts. This similarity is computed by taking into account a user’s explicit browsing behavior, semantic knowledge of concepts, and synonyms of term-based vectors extracted from the WordNet API. A set of experiments using a case study from a transport service domain validates the effectiveness of the proposed approach and demonstrates promising results.

Information generated within the World Wide Web is increasing at huge rate and user's access for their own interested work. Ambiguous queries and lower ability of user to precisely express what they need have been one of the challenging obstacles in improving search results. It is obvious that the current search engines retrieve results are sometimes not of user relevance due to keyword based search, so to fill the gap between the user interest and retrieved search results, personalized web search needs to be evolved. For example, a biologist and a programmer may use the same query “mouse” with different search context, but the search systems would return same results. Again still there is need to customize the search results by re-ranking the retrieved results incorporating the user interests. In this paper we have presents various approaches to personalize web search through user modelling by analyzing semantic data.

There is a growing concern about the extent to which algorithmic personalization limits people's exposure to diverse viewpoints, thereby creating “filter bubbles” or “echo chambers.” Prior research on web search personalization has mainly reported location-based personalization of search results. In this paper, we investigate whether web search results are personalized based on a user's browsing history, which can be inferred by search engines via third-party tracking. Specifically, we develop a “sock puppet” auditing system in which a pair of fresh browser profiles, first, visits web pages that reflect divergent political discourses and, second, executes identical politically oriented Google News searches. Comparing the search results returned by Google News for distinctly trained browser profiles, we observe statistically significant personalization that tends to reinforce the presumed partisanship.

A fundamental problem with current Web search technology is that in the absence of any additional information, the same query provided by two different searchers will produce the same set of search results, even if the information needs of the searchers are different. Web search personalization has been proposed as a solution to this problem, whereby the interests and preferences of individual users are modelled and used to affect the outcomes of their subsequent searches. A common approach is to generate vector-based models of searchers' interests, and re-rank the search results based on the similarity of the documents to these models. In this paper, a novel approach is proposed to automatically identify and re-weight significant dimensions in vector-based models in order to improve the personalized order of Web search results. This approach is inspired by Luhn's model of term importance, which is rooted in Zipf's Laws. Evaluations with a set of ambiguous queries illustrate the effectiveness of this approach.