Actively engaging learners with learning materials has been shown to be very important in the Search as Learning (SAL) setting. One active reading strategy relies on asking so-called adjunct questions, i.e., manually curated questions geared towards essential concepts of the target material. However, manual question creation is impractical given the vast online content. Recent research has explored the effects of Automatic Question Generation (AQG) on aiding human learning. These studies have primarily focused on user studies in controlled online reading scenarios with limited documents. However, the impacts of adjunct questions on learning in the SAL setting, which involves learning through web searching, are not yet well understood. This paper addresses this gap by conducting a user study with automatically generated adjunct questions integrated into the reading interface built on top of a search system. We conducted a between-subjects user study (N = 144) to investigate the incorporation of automatically generated adjunct questions on participants' learning. We employed three different question generation strategies as well as a control condition: (i) synthesis questions; (ii) factoid questions targeting random text spans; and (iii) factoid questions targeting terms and phrases relevant to the information need at hand. We present four major findings: (i) participants who received adjunct questions exhibited significantly more fine-grained reading behaviour, such as longer document dwell time and more scrolls, than those without adjunct questions. However, adjunct questions' influence on learning outcomes depends on the AQG strategy. (ii) Question types significantly influence participants' reading behaviour. (iii) The adjunct questions' target spans significantly influence learning outcomes. Lastly, (iv) participants' prior knowledge levels affect adjunct questions' effects on their learning outcomes and their reaction to different AQG strategies. Our findings have significant design implications for learning-oriented search systems. The data and code is available at https://github.com/zpeide/AQG-AdjunctQuestions.

@en

Advertisements, sponsored links, clickbait, in-house recommendations and similar elements pervasively shroud featured content. Such elements vie for people's attention, potentially distracting people from their task at hand. The effects of such "distractors"is likely to increase people's cognitive workload and reduce their performance as they need to work harder to discern the relevant from non-relevant. In this paper, we investigate how people of varying cognitive abilities (measured using Perceptual Speed and Cognitive Failure instruments) are affected by these different types of distractions when completing search tasks. We performed a crowdsourced within-subjects user study, where 102 participants completed four search tasks using our news search engine over four different interface conditions: (i) one with no additional distractors; (ii) one with advertisements; (iii) one with sponsored links; and (iv) one with in-house recommendations. Our results highlight a number of important trends and findings. Participants perceived the interface condition without distractors as significantly better across numerous dimensions. Participants reported higher satisfaction, lower workload, higher topic recall, and found it easier to concentrate. Behaviourally, participants issued queries faster and clicked results earlier when compared to the interfaces with distractors. When using the interfaces with distractors, one in ten participants clicked on a distractor - and despite engaging with a distractor for less than twenty seconds, their task time increased by approximately two minutes. We found that the effects were magnified depending on cognitive abilities - with a greater impact of distractors on participants with lower perceptual speed, and for those with a higher propensity of cognitive failures. Distractors - regardless of their type - have negative consequences on a user's search experience and performance. As a consequence, interfaces containing visually distracting elements are creating poorer search experiences due to the "distractor tax"being placed on people's limited attention.

@en

A number of learned sparse and dense retrieval approaches have recently been proposed and proven effective in tasks such as passage retrieval and document retrieval. In this paper we analyze with a replicability study if the lessons learned generalize to the retrieval of responses for dialogues, an important task for the increasingly popular field of conversational search. Unlike passage and document retrieval where documents are usually longer than queries, in response ranking for dialogues the queries (dialogue contexts) are often longer than the documents (responses). Additionally, dialogues have a particular structure, i.e. multiple utterances by different users. With these differences in mind, we here evaluate how generalizable the following major findings from previous works are: (F1) query expansion outperforms a no-expansion baseline; (F2) document expansion outperforms a no-expansion baseline; (F3) zero-shot dense retrieval underperforms sparse baselines; (F4) dense retrieval outperforms sparse baselines; (F5) hard negative sampling is better than random sampling for training dense models. Our experiments (https://github.com/Guzpenha/transformer_rankers/tree/full_rank_retrieval_dialogues.)—based on three different information-seeking dialogue datasets—reveal that four out of five findings (F2–F5) generalize to our domain.

@en

The creation of relevance assessments by human assessors (often nowadays crowdworkers) is a vital step when building IR test collections. Prior works have investigated assessor quality & behaviour, and tooling to support assessors in their task. We have few insights though into the impact of a document's presentation modality on assessor efficiency and effectiveness. Given the rise of voice-based interfaces, we investigate whether it is feasible for assessors to judge the relevance of text documents via a voice-based interface. We ran a user study (n = 49) on a crowdsourcing platform where participants judged the relevance of short and long documents-sampled from the TREC Deep Learning corpus-presented to them either in the text or voice modality. We found that: (i) participants are equally accurate in their judgements across both the text and voice modality; (ii) with increased document length it takes participants significantly longer (for documents of length > 120 words it takes almost twice as much time) to make relevance judgements in the voice condition; and (iii) the ability of assessors to ignore stimuli that are not relevant (i.e., inhibition) impacts the assessment quality in the voice modality-assessors with higher inhibition are significantly more accurate than those with lower inhibition. Our results indicate that we can reliably leverage the voice modality as a means to effectively collect relevance labels from crowdworkers.

@en

Natural Language Interfaces to Databases (NLIDB), also known as Text-to-SQL models, enable users with different levels of knowledge in Structured Query Language (SQL) to access relational databases without any programming effort. By translating natural languages into SQL query, not only do NLIDBs minimize the burden of memorizing the schema of databases and writing complex SQL queries, but they also allow non-experts to acquire information from databases in natural languages. However, existing NLIDBs largely fail to translate natural languages to SQL when they are complex, preventing them from being deployed in real-world scenarios and generalizing across unseen complex databases. In this paper, we explored the feasibility of decomposing complex user questions into multiple sub-questions - each with a reduced complexity - as a means to circumvent the problem of complex SQL generation. We investigated the feasibility of decomposing complex user questions in a manner that each sub-question is simple enough for existing NLIDBs to generate correct SQL queries, using non-expert crowd workers in juxtaposition with SQL experts. Through an empirical study on an NLIDB benchmark dataset, we found that crowd-powered decomposition of complex user questions led to an accuracy boost of an existing Text-to-SQL pipeline from 30% to 59% (96% accuracy boost). Similarly, decomposition by SQL experts resulted in boosting the accuracy to 76% (153% accuracy boost). Our findings suggest that crowd-powered decomposition can be a scalable alternative to producing the training data necessary to build machine learning models that can automatically decompose complex user questions, thereby improving Text-to-SQL pipelines.

@en

Heavily pre-trained transformers for language modeling, such as BERT, have shown to be remarkably effective for Information Retrieval (IR) tasks, typically applied to re-rank the results of a first-stage retrieval model. IR benchmarks evaluate the effectiveness of retrieval pipelines based on the premise that a single query is used to instantiate the underlying information need. However, previous research has shown that (I) queries generated by users for a fixed information need are extremely variable and, in particular, (II) neural models are brittle and often make mistakes when tested with modified inputs. Motivated by those observations we aim to answer the following question: how robust are retrieval pipelines with respect to different variations in queries that do not change the queries’ semantics? In order to obtain queries that are representative of users’ querying variability, we first created a taxonomy based on the manual annotation of transformations occurring in a dataset (UQV100) of user-created query variations. For each syntax-changing category of our taxonomy, we employed different automatic methods that when applied to a query generate a query variation. Our experimental results across two datasets for two IR tasks reveal that retrieval pipelines are not robust to these query variations, with effectiveness drops of ≈ 20 % on average. The code and datasets are available at https://github.com/Guzpenha/query_variation_generators.

@en

The Search Engine Results Page (SERP) has evolved significantly over the last two decades, moving away from the simple ten blue links paradigm to considerably more complex presentations that contain results from multiple verticals and granularities of textual information. Prior works have investigated how user interactions on the SERP are influenced by the presence or absence of heterogeneous content (e.g., images, videos, or news content), the layout of the SERP (\emphlist vs. grid layout), and task complexity. In this paper, we reproduce the user studies conducted in prior works---specifically those of~\citetarguello2012task and~\citetsiu2014first ---to explore to what extent the findings from research conducted five to ten years ago still hold today as the average web user has become accustomed to SERPs with ever-increasing presentational complexity. To this end, we designed and ran a user study with four different SERP interfaces:(i) ~\empha heterogeneous grid ;(ii) ~\empha heterogeneous list ;(iii) ~\empha simple grid ; and(iv) ~\empha simple list. We collected the interactions of $41$ study participants over $12$ search tasks for our analyses. We observed that SERP types and task complexity affect user interactions with search results. We also find evidence to support most (6 out of 8) observations from~\citearguello2012task,siu2014first indicating that user interactions with different interfaces and to solve tasks of different complexity have remained mostly similar over time.@en

In this work, we address the information overload issue that learners in Massive Open Online Courses (MOOCs) face when attempting to close their knowledge gaps via the use of MOOC discussion forums. To this end, we investigate the recommendation of one-minute-resolution video clips given the textual similarity between the clips’ transcripts and MOOC discussion forum entries. We first create a large-scale dataset from Khan Academy video transcripts and their forum discussions. We then investigate the effectiveness of applying pre-trained transformers-based neural retrieval models to rank video clips in response to a forum discussion. The retrieval models are trained with supervised learning and distant supervision to effectively leverage the unlabeled data—which accounts for more than 80% of all available data. Our experimental results demonstrate that the proposed method is effective for this task, by outperforming a standard baseline by 0.208 on the absolute change in terms of precision.

@en

Question generation (QG) approaches based on large neural models require (i) large-scale and (ii) high-quality training data. These two requirements pose difficulties for specific application domains where training data is expensive and difficult to obtain. The trained QG models' effectiveness can degrade significantly when they are applied on a different domain due to domain shift. In this paper, we explore an unsupervised domain adaptation approach to combat the lack of training data and domain shift issue with domain data selection and self-training. We first present a novel answer-aware strategy for domain data selection to select data with the most similarity to a new domain. The selected data are then used as pseudo in-domain data to retrain the QG model. We then present generation confidenceguided self-training with two generation confidence modeling methods: (i) generated questions' perplexity and (ii) the fluency score. We test our approaches on three large public datasets with different domain similarities, using a transformer-based pre-trained QG model. The results show that our proposed approaches outperform the baselines, and show the viability of unsupervised domain adaptation with answer-aware data selection and self-training on the QG task. The code is available at https://github.com/zpeide/transfer_qg.

@en

Complex search tasks—such as those from the Search as Learning (SAL) domain—often result in users developing an information need composed of several aspects. However, current models of searcher behaviour assume that individuals have an atomic need, regardless of the task. While these models generally work well for simpler informational needs, we argue that searcher models need to be developed further to allow for the decomposition of a complex search task into multiple aspects. As no searcher model yet exists that considers both aspects and the SAL domain, we propose, by augmenting the Complex Searcher Model (CSM), the Subtopic Aware Complex Searcher Model (SACSM)—modelling aspects as subtopics to the user’s need. We then instantiate several agents (i.e., simulated users), with different subtopic selection strategies, which can be considered as different prototypical learning strategies (e.g., should I deeply examine one subtopic at a time, or shallowly cover several subtopics?). Finally, we report on the first large-scale simulated analysis of user behaviours in the SAL domain. Results demonstrate that the SACSM, under certain conditions, simulates user behaviours accurately.

@en

Quality control is essential for creating extractive question answering (EQA) datasets via crowdsourcing. Aggregation across answers, i.e. word spans within passages annotated, by different crowd workers is one major focus for ensuring its quality. However, crowd workers cannot reach a consensus on a considerable portion of questions. We introduce a simple yet effective answer aggregation method that takes into account the relations among the answer, question, and context passage. We evaluate answer quality from both the view of question answering model to determine how confident the QA model is about each answer and the view of the answer verification model to determine whether the answer is correct. Then we compute aggregation scores with each answer’s quality and its contextual embedding produced by pre-trained language models. The experiments on a large real crowdsourced EQA dataset show that our framework outperforms baselines by around 16% on precision and effectively conduct answer aggregation for extractive QA task.@en

Models developed to simulate user interactions with search interfaces typically do not consider the visual layout and presentation of a Search Engine Results Page (SERP). In particular, the position and size of interfacewidgets ---such as entity cards and query suggestions---are usually considered a negligible constant. In contrast, in this work, we investigate the impact of widget positioning on user behaviour. To this end, we focus on one specific widget: the Query History Widget (QHW). It allows users to see (and thus reflect) on their recently issued queries. We build a novel simulation model based on Search Economic Theory (SET) that considers how users behave when faced with such a widget by incorporating its positioning on the SERP. We derive five hypotheses from our model and experimentally validate them based on user interaction data gathered for an ad-hoc search task, run across five different placements of the \qhw on the SERP. We find partial support for three of the five hypotheses, and indeed observe that a widget's location has a significant impact on search behaviour.@en

Entity cards are a common occurrence in today's web Search Engine Results Pages (SERPs). SERPs provide information on a complex information object in a structured manner. Typically, they combine data from several search verticals. They have been shown to: (i) increase users' engagement with the SERP; and (ii) improve decision making for certain types of searches (such as health searches). In this paper, we investigate whether the benefits of showing entity cards also extend to the Search as Learning (SAL) domain. Do learners learn more when entity cards are present on the SERP? To answer this question, we designed a series of learning-oriented search tasks (with a minimum search time of 15 minutes), and conducted a crowdsourced Interactive Information Retrieval (IIR) user study (N=144) with four interface conditions: (i) a control with no entity cards; (ii) displaying relevant entity cards; (iii) displaying somewhat relevant entity cards; and (iv) displaying non-relevant entity cards. Our results show that (i) entity cards do not have an effect on participants' learning, but (ii) they do significantly impact participants' search behaviours across a range of dimensions (such as the dwell time and search session duration).@en

Logging user interactions is fundamental to capturing and subsequently analysing user behaviours in the context of web-based Interactive Information Retrieval (IIR). However, logging is often implemented within experimental apparatus in a piecemeal fashion, leading to incomplete or noisy data. To address these issues, we present the LogUI logging framework. We use (now ubiquitous) contemporary web technologies to provide an easy-to-use yet powerful framework that can capture virtually any user interaction on a webpage. LogUI removes many of the complexities that must be considered for effective interaction logging.@en

Prior work in education research has shown that various active reading strategies, notably highlighting and note-taking, benefit learning outcomes. Most of these findings are based on observational studies where learners learn from a single document. In a Search as Learning (SAL) context where learners have to iteratively scan and explore a large number of documents to address their learning objective, the effect of these active reading strategies is largely unexplored. To address this research gap, we carried out a crowd-sourced user study, and explored the effects of different highlighting and note-taking strategies on learning during a complex, learning-oriented search task. Out of five hypotheses derived from the education literature we could confirm three in the SAL context. Our findings have important design implications on aiding learning through search. Learners can benefit from search interfaces equipped with active reading tools—but some learning strategies employing these tools are more effective than others. (This research has been supported by DDS (Delft Data Science) and NWO projects SearchX (639.022.722) and Aspasia (015.013.027).)

@en

An introduction to the special issue on conversational search and recommendation is presented in this article. While conversational search and recommendation has roots in early Information Retrieval (IR) research, the recent advances in automatic voice recognition and conversational agents have created increasing interest in this area. In recent years, the IR and related communities have witnessed a number of major contributions to the field of conversational search and recommendation. They include but are not limited to conversational search conceptualization. The growing body of work in this area has been supplemented by an increasing number of recent seminars.

@en

Active reading strategies - -such as content annotations (through the use of highlighting and note-taking, for example) - -have been shown to yield improvements to a learner's knowledge and understanding of the topic being explored. This has been especially notable in long and complex learning endeavours. With web search engines nowadays used as the primary gateway for learners (or users) to find content that helps them realise their learning goals, they are often poorly equipped with the necessary tools to aid in sense-making, an important aspect of theSearch as Learning (SAL) process. Within theInformation Retrieval (IR) community, research efforts have explored ways to keep track of users' search context by providing a notepad-like interface for the collection of relevant articles, and aid them during the exploratory search process. However, these studies did not explicitly measure the effect that such tools have on knowledge and understanding during a complex, learning-oriented search task. In this paper, we address this research gap by carrying out an InteractiveIR experiment with highlighting and note-taking tools built into the search interface. We conducteda crowdsourced between-subjects study (N=115), where participants were assigned to one of four conditions: (i) control (a standard web search interface); (ii) high (highlighting enabled);(iii) note (note-taking enabled); and (iv) highnote (both highlighting and note-taking enabled). We assess participants' learning with a recall-oriented vocabulary learning task, and a cognitively more taxing essay writing task. We find that(i) active reading tools do not aid in the vocabulary learning task. However,(ii) participants in high covered 34% more subtopics, and participants in note covered 34% more facts in their essays when compared to control. Furthermore, (iii) we observed that incorporating active learning tools significantly changed the search behaviour of participants across a number of measures. This is the first work that sheds light on the effect of active reading tools on the SAL process, with important design implications for learning-oriented search systems.

@en

We study Label Smoothing (LS), a widely used regularization technique, in the context of neural learning to rank (L2R) models. LS combines the ground-truth labels with a uniform distribution, encouraging the model to be less confident in its predictions. We analyze the relationship between the non-relevant documents—specifically how they are sampled—and the effectiveness of LS, discussing how LS can be capturing “hidden similarity knowledge” between the relevant and non-relevant document classes. We further analyze LS by testing if a curriculum-learning approach, i.e., starting with LS and after a number of iterations using only ground-truth labels, is beneficial. Inspired by our investigation of LS in the context of neural L2R models, we propose a novel technique called Weakly Supervised Label Smoothing (WSLS) that takes advantage of the retrieval scores of the negative sampled documents as a weak supervision signal in the process of modifying the ground-truth labels. WSLS is simple to implement, requiring no modification to the neural ranker architecture. Our experiments across three retrieval tasks—passage retrieval, similar question retrieval and conversation response ranking—show that WSLS for pointwise BERT-based rankers leads to consistent effectiveness gains. The source code is available at https://github.com/Guzpenha/transformer_rankers/tree/wsls.@en

Web search engines are today considered to be the primary tool to assist and empower learners in finding information relevant to their learning goals- be it learning something new, improving their existing skills, or just fulfilling a curiosity. While several approaches for improving search engines for the learning scenario have been proposed (e.g. a specific ranking function), instructional scaffolding (or simply scaffolding)-a traditional learning support strategy-has not been studied in the context of search as learning, despite being shown to be effective for improving learning in both digital and traditional learning contexts. When scaffolding is employed, instructors provide learners with support throughout their autonomous learning process. We hypothesize that the usageof scaffolding techniques within a search system can be an effective way to help learners achieve their learning objectives whilst searching. As such, this paper investigates the incorporation of scaffolding into a search system employing three different strategies (as well as a control condition): (i) AQe, the automatic expansion of user queries with relevant subtopics; (ii) CURATEDsc, the presenting of a manually curated static list of relevant subtopics on the search engine result page; and (iii) FEEDBACKsc, which projects real-time feedback about a user's exploration of the topic space on top of the CURATEDsc visualization. To investigate the effectiveness of these approaches withrespect to human learning, we conduct a user study (N=126) where participants were tasked with searching and learning about topics such as genetically modified organisms. We find that (i) the introduction of the proposed scaffolding methods in the proposed topics does not significantly improve learning gains. However, (ii) it does significantly impact search behavior. Furthermore, (iii) immediate feedback of the participants' learning (FEEDBACKsc) leads to undesirable user behavior, with participants seemingly focusing on the feedback gauges instead of learning.

@en

Question generation systems aim to generate natural language questions that are relevant to a given piece of text, and can usually be answered by just considering this text. Prior works have identified a range of shortcomings (including semantic drift and exposure bias) and thus have turned to the reinforcement learning paradigm to improve the effectiveness of question generation. As part of it, different reward functions have been proposed. As typically these reward functions have been empirically investigated in different experimental settings (different datasets, models and parameters) we lack a common framework to fairly compare them. In this paper, we first categorize existing rewards systematically. We then provide such a fair empirical evaluation of different reward functions (including three we propose here for QG) in a common framework. We find rewards that model answerability to be the most effective.

@en