Anytime Ranking on Document-Ordered Indexes

Inverted indexes continue to be a mainstay of text search engines, allowing efficient querying of large document collections. While there are a number of possible organizations, document-ordered indexes are the most common, since they are amenable to various query types, support index updates, and allow for efficient dynamic pruning operations. One disadvantage with document-ordered indexes is that high-scoring documents can be distributed across the document identifier space, meaning that index traversal algorithms that terminate early might put search effectiveness at risk. The alternative is impact-ordered indexes, which primarily support top- disjunctions but also allow for anytime query processing, where the search can be terminated at any time, with search quality improving as processing latency increases. Anytime query processing can be used to effectively reduce high-percentile tail latency that is essential for operational scenarios in which a service level agreement (SLA) imposes response time requirements. In this work, we show how document-ordered indexes can be organized such that they can be queried in an anytime fashion, enabling strict latency control with effective early termination. Our experiments show that processing document-ordered topical segments selected by a simple score estimator outperforms existing anytime algorithms, and allows query runtimes to be accurately limited to comply with SLA requirements.

Starting 2022 with a new feature: full-text search for the Front Matter blog

Fresh into 2022, the Front Matter blog today is launching an important new feature: full-text search of all blog posts. An example query would be for reference manager:As the Front Matter blog has a lot of posts about reference managers, ...

Taking “Use Case” Inventory of Available Open Shared Visuals for Teaching and Learning From Searches in the Federated Creative Commons Search (Old)

In instructional design, there are a number of common “use cases” for acquiring open-source shared visuals and images: breaking up gray text, driving attention, sparking the imagination, illustrating concepts, providing examples, explaining phenomena, representing reality, depicting models, and others. The instating of licensure and open-source releases has meant that there are literally hundreds of millions of such visuals available online, with varying levels of releases (with variations on the following dimensions: editability, [non]crediting, [non]commercial usages, [non]required sharing, all the way up to full release into the public domain with no restrictions). The federated Creative Commons Search (old) enables exploration and acquisition across a range of web-based platforms for digital images based on text search. When pursuing actual images for particular usage, the abundance of shared imagery suddenly becomes small-set and limited. This work explores this phenomenon and provides some ideas for mitigation.

Intelligent Resume Retrieval Based on Lucence

With the development of Internet, the electronic resume has gradually replaced the paper one. It is the basic requirement of recruitment for enterprises to retrieve the talent information that fulfills the requirement quickly and without omission.Based on the framework of SpringBoot and Lucence full-text search engine, this paper implements a resume intelligent filtering algorithm, which improves the query speed of the system by establishing an index database. At the same time,the scoring function improves the accuracy of the filtering results, reduces the pressure of high concurrency of the database, improves the work efficiency of the Human Resources Department, and avoids the talent loss.

An efficient and scalable search engine for models

Search engines extract data from relevant sources and make them available to users via queries. A search engine typically crawls the web to gather data, analyses and indexes it and provides some query mechanism to obtain ranked results. There exist search engines for websites, images, code, etc., but the specific properties required to build a search engine for models have not been explored much. In the previous work, we presented MAR, a search engine for models which has been designed to support a query-by-example mechanism with fast response times and improved precision over simple text search engines. The goal of MAR is to assist developers in the task of finding relevant models. In this paper, we report new developments of MAR which are aimed at making it a useful and stable resource for the community. We present the crawling and analysis architecture with which we have processed about 600,000 models. The indexing process is now incremental and a new index for keyword-based search has been added. We have also added a web user interface intended to facilitate writing queries and exploring the results. Finally, we have evaluated the indexing times, the response time and search precision using different configurations. MAR has currently indexed over 500,000 valid models of different kinds, including Ecore meta-models, BPMN diagrams, UML models and Petri nets. MAR is available at http://mar-search.org.

Pengalaman Orang Tua dalam Merawat Anak Berkebutuhan Khusus : Literature Review

Introduction: The presence of a child in the family is certainly very encouraging for parents. However, it is different from parents who have children with special needs. Children with special needs need different treatment from other children. This of course raises different experiences for each parent in their care. Objective: This study was to determine the psychosocial experience of parents in caring for children with special needs. Method: The method used in this paper is a literature review. With library sources, namely journal articles published in the 2020-2021 period which are full text. Search for journal articles using the Google Schoolar database with the keyword experience; parent; nurse; the child with special needs. Results: This study found 1,500 journal articles which the researchers then took according to the specified criteria, obtained as many as 10 articles. 10 articles reviewed by researchers found 3 journal articles on experiences of parents who can accept the condition of children with special needs and 7 articles found experiences of parents who have psychosocial problems in caring for children with special needs. Conclusion: This literature review found that the experience of parents in caring for children with special needs is divided into two where there are parents who can accept their child's condition sincerely and parents who experience psychosocial problems in the care of children with special needs such as stress, inferiority, shock, rejection, etc. How parents respond to their children with special needs is influenced by many factors such as age, environment, knowledge, etc.

Sustainable Development Indicators—Untapped Tools for Sustainability and STEM Education: An Analysis of a Popular Czech Educational Website

Environmental education has been included in Czech curricula since the 1980s, albeit without clear evidence of education for sustainable development (SD), which addresses complex socio-economic issues using SD indicators (SDIs), such as charts, single numbers, tables, maps, and (interactive) images. However, understanding such a comprehensive topic requires developing basic mathematical knowledge and skills. In this study, we aimed to analyse the nature, quality, and availability of teaching materials for SD, primarily using SDIs, which could be applied by Czech teachers. For this purpose, we performed a qualitative and basic quantitative content analysis of several descriptors of documents retrieved from a website for teachers, provided by the National Pedagogical Institute of the Czech Republic. A full-text search identified 1376 records, which were analyzed for SD pillars and SDIs. Our results showed that most records (95%) do not contain SDIs in teaching materials. Only 59 records mentioned (128) SDIs, mostly covering the environmental pillar, 26 of which contain a single SDI. The most frequent issues were waste production, treatment, savings, water parameters, and energy consumption. Mathematical skills were used in 56 SDIs, primarily for evaluating data sets and quantitative expressions of an amount. Overall, only a small number of SDIs are used in education for SD, economic and social SDIs are in the minority, and the STEM potential remains untapped.

Full Text Search Setup on a Website

ntroduction. When implementing the search for text fragments on the site, approaches are used that are different in complexity and performance. There is also a sequence of related tasks: choosing a text indexing option, sending a text for indexing, selecting texts for indexing specifically from the CMS database, choosing a search engine, and others. These approaches do not always provide satisfactory search results. Purpose. The purpose of the article is to the description of existing solutions for full-text search on a website, their advantages, and disadvantages. Development of a full-text search algorithm using the Elasticsearch system. Methods. Analysis of approaches to the implementation of full-text search on a website, varying in complexity and performance. Identification of flaws and vulnerabilities in more primitive approaches and the development of more advanced and complex algorithms that eliminate the identified deficiencies. Step-by-step implementation of full-text search using third-party systems. Results. A method for implementing full-text search using Elasticsearch is described. The advantage of the new approach is the asynchronous sending of the page content and its address to a specific service responsible for communication with Elasticsearch. This allows you not to block the normal work with the CMS and not depend on the availability of the indexing service. The approach described in the article is flexible and adaptable for various website architectures. Asynchronous processing of indexing requests ensures high query execution speed and system fault tolerance. Conclusions. The article discusses various approaches to implementing full-text search on a website, their advantages and disadvantages. Based on the analysis, a more flexible and universal approach to the implementation of a full-text search system has been developed. A solution is proposed with step-by-step implementation and setup of advanced full-text search using Elasticsearch.

Table to text generation with accurate content copying

Generating fluent, coherent, and informative text from structured data is called table-to-text generation. Copying words from the table is a common method to solve the “out-of-vocabulary” problem, but it’s difficult to achieve accurate copying. In order to overcome this problem, we invent an auto-regressive framework based on the transformer that combines a copying mechanism and language modeling to generate target texts. Firstly, to make the model better learn the semantic relevance between table and text, we apply a word transformation method, which incorporates the field and position information into the target text to acquire the position of where to copy. Then we propose two auxiliary learning objectives, namely table-text constraint loss and copy loss. Table-text constraint loss is used to effectively model table inputs, whereas copy loss is exploited to precisely copy word fragments from a table. Furthermore, we improve the text search strategy to reduce the probability of generating incoherent and repetitive sentences. The model is verified by experiments on two datasets and better results are obtained than the baseline model. On WIKIBIO, the result is improved from 45.47 to 46.87 on BLEU and from 41.54 to 42.28 on ROUGE. On ROTOWIRE, the result is increased by 4.29% on CO metric, and 1.93 points higher on BLEU.

Prediction Candidates and Political Parties in the Presidential Election 2024 in Indonesia Based on Twitter

Twitter is a popular platform on social media that is used to predict presidential candidates and political parties who will contest in the presidential election. This study uses a quantitative approach with descriptive content analysis. This approach describes the details of a text or message related to discussions and information on the Twitter social network in the 2024 presidential election. The research subjects are Twitter social media users based on the involvement of Twitter users in the 2024 presidential election discourse in Indonesia. The data is obtained from Twitter with Twitter Search focusing on the keyword “Pilpres 2024”. The analytical tool used is Nvivo 12 Plus software with Word Frequency Query and Text Search Query analysis features. This study predicts that candidates with a strong chance as official candidates are Anies Baswedan, Prabowo, and Ganjar Pranowo. The mapping of political parties indicates that there will be political contestation between nationalist parties and religious-based parties in the 2024 presidential election.