A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Filters
Unfairness towards subjective opinions in Machine Learning
[article]
2019
arXiv
pre-print
Preserved Fulltext
Despite the high interest for Machine Learning (ML) in academia and industry, many issues related to the application of ML to real-life problems are yet to be addressed. Here we put forward one limitation which arises from a lack of adaptation of ML models and datasets to specific applications. We formalise a new notion of unfairness as exclusion of opinions. We propose ways to quantify this unfairness, and aid understanding its causes through visualisation. These insights into the functioning
arXiv:1911.02455v1
fatcat:5fjqhfk5orexrhsoxzqlxhn6la
more »
... f ML-based systems hint at methods to mitigate unfairness.
Metadata Representations for Queryable ML Model Zoos
[article]
2022
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
Machine learning (ML) practitioners and organizations are building model zoos of pre-trained models, containing metadata describing properties of the ML models and datasets that are useful for reporting, auditing, reproducibility, and interpretability purposes. The metatada is currently not standardised; its expressivity is limited; and there is no interoperable way to store and query it. Consequently, model search, reuse, comparison, and composition are hindered. In this paper, we advocate for
arXiv:2207.09315v1
fatcat:okbik52a2na57oiia7wfpta7ki
more »
... standardized ML model meta-data representation and management, proposing a toolkit supported to help practitioners manage and query that metadata.
Modeling Search Computing Applications
[chapter]
2010
Lecture Notes in Computer Science
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2018; you can also visit the original URL.
The file type is application/pdf.
Search Computing defines a new class of applications, which enable end users to perform exploratory search processes over multi-domain data sources available on the Web. These applications exploit suitable models, supported by a framework, that make it possible for expert users to configure the data sources to be searched and the interfaces for query submission and result visualization, by using for such source and interface configurations mash-up tools which do not require programming. This
doi:10.1007/978-3-642-16985-4_6
fatcat:bwrd643ubvbcxpnfdkzqhhqcsy
more »
... er presents Search Computing design process and developer roles, together with the conceptual models that represent the foundation of a model-driven approach to Search Computing.
Inferring Student Attention with ASQ
[chapter]
2016
Lecture Notes in Computer Science
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2019; you can also visit the original URL.
The file type is application/pdf.
ASQ is a Web application for broadcasting and tracking interactive presentations, which can be used to support active learning pedagogies during lectures, labs and exercise sessions. Students connect their smartphones, tablets or laptops to receive the current slide as it is being explained by the teacher. Slides can include interactive teaching elements (usually questions of different forms). In contrast to other existing platforms, ASQ does not only collect, aggregate and visualize the
doi:10.1007/978-3-319-45153-4_23
fatcat:y5qe2qjhl5gfxpdelkczu6a7di
more »
... in real-time, it also supports the data analytics in the classroom paradigm by providing the teacher with a real-time analysis of student behaviour during the entire session. One vital aspect of student behaviour is (in)attention and in this paper we discuss how we infer -in real-time -student attention based on log traces ASQ collects.
Towards an Ontological Representation of Services in Search Computing
[chapter]
2011
Lecture Notes in Computer Science
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2011; you can also visit the original URL.
The file type is application/pdf.
In the Search Computing project, Web services are modeled by the Semantic Resource Framework (SRF). In this article, we argue that the SRF could benefit from ontological concepts borrowed from the Semantic Web. We first present the knowledge representation used in the Semantic Web, notably in the YAGO ontology [14] . We show how this model is used in the ANGIE system [12] to represent Web Services in conjunction with YAGO. We draw parallels to the Service Mart [3] model used in SeCo. We propose
doi:10.1007/978-3-642-19668-3_9
fatcat:owosix2blfbgrp4anfkyi22tpy
more »
... a symbiosis of the two models, discussing the challenges and advantages that come with the integrated model.
A Model-Driven Approach for Crowdsourcing Search
2012
The Web Conference
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
Even though search systems are very efficient in retrieving world-wide information, they can not capture some peculiar aspects and features of user needs, such as subjective opinions and recommendations, or information that require local or domain specific expertise. In this kind of scenario, the human opinion provided by an expert or knowledgeable user can be more useful than any factual information retrieved by a search engine. In this paper we propose a model-driven approach for the
dblp:conf/www/BozzonBM12
fatcat:tgbxxfhhxrgjrl4s6k2dnmpvju
more »
... tion of crowd-search tasks, i.e. activities where real people -in real time -take part to the generalized search process that involve search engines. In particular we define two models: the "Query Task Model", representing the metamodel of the query that is submitted to the crowd and the associated answers; and the "User Interaction Model", which shows how the user can interact with the query model to fulfill her needs. Our solution allows for a top-down design approach, from the crowd-search task design, down to the crowd answering system design. Our approach also grants automatic code generation thus leading to quick prototyping of search applications based on human responses collected over social networking or crowdsourcing platforms.
Searching Repositories of Web Application Models
[chapter]
2010
Lecture Notes in Computer Science
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2018; you can also visit the original URL.
The file type is application/pdf.
Project repositories are a central asset in software development, as they preserve the technical knowledge gathered in past development activities. However, locating relevant information in a vast project repository is problematic, because it requires manually tagging projects with accurate metadata, an activity which is time consuming and prone to errors and omissions. This paper investigates the use of classical Information Retrieval techniques for easing the discovery of useful information
doi:10.1007/978-3-642-13911-6_1
fatcat:za6y2qnsnzh4ve74c244o57x6u
more »
... om past projects. Differently from approaches based on textual search over the source code of applications or on querying structured metadata, we propose to index and search the models of applications, which are available in companies applying Model-Driven Engineering practices. We contrast alternative index structures and result presentations, and evaluate a prototype implementation on real-world experimental data.
Reactive crowdsourcing
2013
Proceedings of the 22nd international conference on World Wide Web - WWW '13
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2013; you can also visit the original URL.
The file type is application/pdf.
An essential aspect for building effective crowdsourcing computations is the ability of "controlling the crowd", i.e. of dynamically adapting the behaviour of the crowdsourcing systems as response to the quantity and quality of completed tasks or to the availability and reliability of performers. Most crowdsourcing systems only provide limited and predefined controls; in contrast, we present an approach to crowdsourcing which provides fine-level, powerful and flexible controls. We model each
doi:10.1145/2488388.2488403
dblp:conf/www/BozzonBCM13
fatcat:xzvtm7xgn5gd7n5xmvbxisehrm
more »
... wdsourcing application as composition of elementary task types and we progressively transform these high level specifications into the features of a reactive execution environment that supports task planning, assignment and completion as well as performer monitoring and exclusion. Controls are specified as active rules on top of data structures which are derived from the model of the application; rules can be added, dropped or modified, thus guaranteeing maximal flexibility with limited effort. We also report on our prototype platform that implements the proposed framework and we show the results of our experimentations with different rule sets, demonstrating how simple changes to the rules can substantially affect time, effort and quality involved in crowdsourcing activities.
Liquid query
2010
Proceedings of the 19th international conference on World wide web - WWW '10
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
In this paper we propose the Liquid Query paradigm, to support users in finding responses to multi-domain queries through exploratory information seeking across structured information sources (Web documents, deep Web data, and personal data repositories), wrapped by means of a uniform notion of search service. Liquid Query aims at filling the gap between generalpurpose search engines, which are unable to find information spanning multiple topics, and domain-specific search systems, which cannot
doi:10.1145/1772690.1772708
dblp:conf/www/BozzonBCF10
fatcat:23m5obav4fhwfewbslyca5r7qu
more »
... go beyond their domain limits. The Liquid Query interface consists of interaction primitives that let users pose questions and explore results spanning over multiple sources incrementally, thus getting closer and closer to the sought information. We demonstrate our approach with a prototype built upon the YQL (Yahoo! Query Language) framework.
Diversification for Multi-domain Result Sets
[chapter]
2012
Lecture Notes in Computer Science
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
Multi-domain search answers to queries spanning multiple entities, like "Find an affordable house in a city with low criminality index, good schools and medical services", by producing ranked sets of entity combinations that maximize relevance, measured by a function expressing the user's preferences. Due to the combinatorial nature of results, good entity instances (e.g., inexpensive houses) tend to appear repeatedly in top-ranked combinations. To improve the quality of the result set, it is
doi:10.1007/978-3-642-31753-8_10
fatcat:csabsrc4ejdynmp3leyxjr4eui
more »
... portant to balance relevance (i.e., high values of the ranking function) with diversity, which promotes different, yet almost equally relevant, entities in the top-k combinations. This paper explores two different notions of diversity for multi-domain result sets, compares experimentally alternative algorithms for the trade-off between relevance and diversity, and performs a user study for evaluating the utility of diversification in multi-domain queries.
Social Smart Meter
2018
Companion of the The Web Conference 2018 on The Web Conference 2018 - WWW '18
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Having a thorough understanding of energy consumption behavior is an important element of sustainability studies. Traditional sources of information about energy consumption, such as smart meter devices and surveys, can be costly to deploy, may lack contextual information or have infrequent updates. In this paper, we examine the possibility of extracting energy consumption-related information from user-generated content. More specifically, we develop a pipeline that helps identify
doi:10.1145/3184558.3186977
dblp:conf/www/0001PB18
fatcat:fdrvskotdraobin7mhuckkpdam
more »
... content in Twitter posts and classify it into four categories (dwelling, food, leisure, and mobility), according to the type of activity performed. We further demonstrate a web-based application -called Social Smart Meter -that implements the proposed pipeline and enables different stakeholders to gain an insight into daily energy consumption behavior, as well as showcase it in case studies involving several world cities.
Efficient Execution of Top-K SPARQL Queries
[chapter]
2012
Lecture Notes in Computer Science
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2019; you can also visit the original URL.
The file type is application/pdf.
Top-k queries, i.e. queries returning the top k results ordered by a user-defined scoring function, are an important category of queries. Order is an important property of data that can be exploited to speed up query processing. State-of-the-art SPARQL engines underuse order, and top-k queries are mostly managed with a materialize-then-sort processing scheme that computes all the matching solutions (e.g. thousands) even if only a limited number k (e.g. ten) are requested. The SPARQL-RANK
doi:10.1007/978-3-642-35176-1_22
fatcat:wtfyzolrvbhopnrvhm3dxfio3i
more »
... is an extended SPARQL algebra that treats order as a first class citizen, enabling efficient split-and-interleave processing schemes that can be adopted to improve the performance of top-k SPARQL queries. In this paper we propose an incremental execution model for SPARQL-RANK queries, we compare the performance of alternative physical operators, and we propose a rank-aware join algorithm optimized for native RDF stores. Experiments conducted with an open source implementation of a SPARQL-RANK query engine based on ARQ show that the evaluation of top-k queries can be sped up by orders of magnitude. 1 SELECT ?product ?of fer (g1(?avgRat1) + g2(?avgRat2) + g3(?price1) AS ?score) 2 WHERE { ?product hasAvgRating1 ?avgRat1 . 3 ?product hasAvgRating2 ?avgRat2 . 4 ?product hasName ?name . 5 ?product hasOffers ?of fer . 6 ?of fer hasPrice ?price1 } 7 ORDER BY DESC(?score) LIMIT 10 Listing 1.1. A top-k SPARQL query that retrieves the best ten offers of products ordered by a function of user ratings and offer price; gi are normalization functions and the bold letters represent the abbreviations used in the following examples P. Cudré-Mauroux et al. (Eds.): ISWC 2012, Part I, LNCS 7649, pp. 344-360, 2012.
On Development Practices for End Users
[chapter]
2011
Lecture Notes in Computer Science
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
The paper discusses some trends in end user programming (EUP) and takes inspiration from the discussions in a panel and in a vertical session on research evaluation within the second Search Computing workshop. We discuss the controversial successes and failures in this field and we elaborate on which facilities could foster adoption of end user programming. We discuss various dimensions of end user programming, including vertical versus horizontal language definition, declarative versus
doi:10.1007/978-3-642-19668-3_18
fatcat:2wii24cjrffv7b4v45i5lwj2vu
more »
... ve approaches. We exemplify our discussion in the realistic scenario of research evaluation by comparing the Search Computing and ResEval approaches.
CitRec 2017
2017
Proceedings of the Eleventh ACM Conference on Recommender Systems - RecSys '17
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2018; you can also visit the original URL.
The file type is application/pdf.
Bozzon, Jie Zhang, and Martha Larson recommendation based techniques to enhance citizens' living experiences. ...
Thus they describe comprehensively citizens' behaviors and their relationships with the environment, providing opportunities for RecSys '17, August 27-31, 2017, Como, Italy Jie Yang, Zhu Sun, Alessandro ...
doi:10.1145/3109859.3109969
dblp:conf/recsys/YangSBZL17
fatcat:w7glpy4gmjfmvn6eccvl53i4ni
Designing Evaluations of Machine Learning Models for Subjective Inference: The Case of Sentence Toxicity
[article]
2019
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Machine Learning (ML) is increasingly applied in real-life scenarios, raising concerns about bias in automatic decision making. We focus on bias as a notion of opinion exclusion, that stems from the direct application of traditional ML pipelines to infer subjective properties. We argue that such ML systems should be evaluated with subjectivity and bias in mind. Considering the lack of evaluation standards yet to create evaluation benchmarks, we propose an initial list of specifications to
arXiv:1911.02471v1
fatcat:cjac3tgjmjgoxnhswqcocjemcu
more »
... prior to creating evaluation datasets, in order to later accurately evaluate the biases. With the example of a sentence toxicity inference system, we illustrate how the specifications support the analysis of biases related to subjectivity. We highlight difficulties in instantiating these specifications and list future work for the crowdsourcing community to help the creation of appropriate evaluation datasets.
« Previous
Showing results 1 — 15 out of 148 results