Query Auto-completion Research Articles

BWBEV: A Bitwise Query Processing Algorithm for Approximate Prefix Search

We tackle the challenge of conducting an approximate prefix search within datasets of strings. We explore using a bit-parallelism technique to compute the edit distance between distinct strings and illustrate its adaptation for an approximate prefix search procedure referred to as BWBEV. This technique employs a unary representation of edit vectors alongside bitwise operations to efficiently update these vectors during the edit distance computation. We also show how to apply our new bit-parallelism technique strategy to online edit distance computation between strings without index structure. Our experiments with BWBEV applied to approximate prefix search for a query autocompletion task revealed a substantial acceleration of over 36% when contrasted against state-of-the-art methods.

Predicting Representations of Information Needs from Digital Activity Context

Information retrieval systems often consider search-session and immediately preceding web-browsing history as the context for predicting users’ present information needs. However, such context is only available when a user’s information needs originate from web context or when users have issued preceding queries in the search session. Here, we study the effect of more extensive context information recorded from users’ everyday digital activities by monitoring all information interacted with and communicated using personal computers. Twenty individuals were recruited for 14 days of 24/7 continuous monitoring of their digital activities, including screen contents, clicks, and operating system logs on Web and non-Web applications. Using this data, a transformer architecture is applied to model the digital activity context and predict representations of personalized information needs. Subsequently, the representations of information needs are used for query prediction, query auto-completion, selected search result prediction, and Web search re-ranking. The predictions of the models are evaluated against the ground truth data obtained from the activity recordings. The results reveal that the models accurately predict representations of information needs improving over the conventional search session and web-browsing contexts. The results indicate that the present practice for utilizing users’ contextual information is limited and can be significantly extended to achieve improved search interaction support and performance.

Discriminator-Enhanced Knowledge-Distillation Networks

Query auto-completion (QAC) serves as a critical functionality in contemporary textual search systems by generating real-time query completion suggestions based on a user’s input prefix. Despite the prevalent use of language models (LMs) in QAC candidate generation, LM-based approaches frequently suffer from overcorrection issues during pair-wise loss training and efficiency deficiencies. To address these challenges, this paper presents a novel framework—discriminator-enhanced knowledge distillation (Dis-KD)—for the QAC task. This framework combines three core components: a large-scale pre-trained teacher model, a lightweight student model, and a discriminator for adversarial learning. Specifically, the discriminator aids in discerning generative-level differences between the teacher and the student models. An additional discriminator score loss is amalgamated with the traditional knowledge-distillation loss, resulting in enhanced performance of the student model. Contrary to the stepwise evaluation of each generated word, our approach assesses the entire generation sequence. This method alleviates the prevalent overcorrection issue in the generation process. Consequently, our proposed framework boasts improvements in model accuracy and a reduction in parameter size. Empirical results highlight the superiority of Dis-KD over established baseline methods, with the student model surpassing the teacher model in QAC tasks for sub-word languages.

FLAG: Towards Graph Query Autocompletion for Large Graphs

Graph query autocompletion (GQAC) takes a user’s graph query as input and generates top-k query suggestions as output, to help alleviate the verbose and error-prone graph query formulation process in a visual interface. To compose a target query with GQAC, the user may iteratively adopt suggestions or manually add edges to augment the existing query. The current state-of-the-art of GQAC, however, focuses on a large collection of small- or medium-sized graphs only. The subgraph features exploited by existing GQAC are either too small or too scarce in large graphs. In this paper, we present Flexible graph query autocompletion for LArge Graphs, called FLAG. We are the first to propose wildcard labels in the context of GQAC, which summarizes query structures that have different labels. FLAG allows augmenting users’ queries with subgraph increments with wildcard labels to form suggestions. To support wildcard-enabled suggestions, a new suggestion ranking function is proposed. We propose an efficient ranking algorithm and extend an index to further optimize the online suggestion ranking. We have conducted a user study and a set of large-scale simulations to verify both the effectiveness and efficiency of FLAG. The results show that the query suggestions saved roughly 50% of mouse clicks and FLAG returns suggestions in few seconds.

Neural methods for effective, efficient, and exposure-aware information retrieval

Neural networks with deep architectures have demonstrated significant performance improvements in computer vision, speech recognition, and natural language processing. The challenges in information retrieval (IR), however, are different from these other application areas. A common form of IR involves ranking of documents---or short passages---in response to keyword-based queries. Effective IR systems must deal with query-document vocabulary mismatch problem, by modeling relationships between different query and document terms and how they indicate relevance. Models should also consider lexical matches when the query contains rare terms---such as a person's name or a product model number---not seen during training, and to avoid retrieving semantically related but irrelevant results. In many real-life IR tasks, the retrieval involves extremely large collections---such as the document index of a commercial Web search engine---containing billions of documents. Efficient IR methods should take advantage of specialized IR data structures, such as inverted index, to efficiently retrieve from large collections. Given an information need, the IR system also mediates how much exposure an information artifact receives by deciding whether it should be displayed, and where it should be positioned, among other results. Exposure-aware IR systems may optimize for additional objectives, besides relevance, such as parity of exposure for retrieved items and content publishers. In this thesis, we present novel neural architectures and methods motivated by the specific needs and challenges of IR tasks. We ground our contributions with a detailed survey of the growing body of neural IR literature [Mitra and Craswell, 2018]. Our key contribution towards improving the effectiveness of deep ranking models is developing the Duet principle [Mitra et al., 2017] which emphasizes the importance of incorporating evidence based on both patterns of exact term matches and similarities between learned latent representations of query and document. To efficiently retrieve from large collections, we develop a framework to incorporate query term independence [Mitra et al., 2019] into any arbitrary deep model that enables large-scale precomputation and the use of inverted index for fast retrieval. In the context of stochastic ranking, we further develop optimization strategies for exposure-based objectives [Diaz et al., 2020]. Finally, this dissertation also summarizes our contributions towards benchmarking neural IR models in the presence of large training datasets [Craswell et al., 2019] and explores the application of neural methods to other IR tasks, such as query auto-completion.

Spoken Conversational Context Improves Query Auto-completion in Web Search

Web searches often originate from conversations in which people engage before they perform a search. Therefore, conversations can be a valuable source of context with which to support the search process. We investigate whether spoken input from conversations can be used as a context to improve query auto-completion. We model the temporal dynamics of the spoken conversational context preceding queries and use these models to re-rank the query auto-completion suggestions. Data were collected from a controlled experiment and comprised conversations among 12 participant pairs conversing about movies or traveling. Search query logs during the conversations were recorded and temporally associated with the conversations. We compared the effects of spoken conversational input in four conditions: a control condition without contextualization; an experimental condition with the model using search query logs; an experimental condition with the model using spoken conversational input; and an experimental condition with the model using both search query logs and spoken conversational input. We show the advantage of combining the spoken conversational context with the Web-search context for improved retrieval performance. Our results suggest that spoken conversations provide a rich context for supporting information searches beyond current user-modeling approaches.

TIPS: Time-aware Personalised Semantic-based query auto-completion

With the rapid growth of the Internet, search engines play vital roles in meeting the users’ information needs. However, formulating information needs to simple queries for canonical users is a problem yet. Therefore, query auto-completion, which is one of the most important characteristics of the search engines, is leveraged to provide a ranked list of queries matching the user’s entered prefix. Although query auto-completion utilises useful information provided by search engine logs, time-, semantic- and context-aware features are still important resources of extra knowledge. Specifically, in this study, a hybrid query auto-completion system called TIPS ( Time-aware Personalised Semantic-based query auto-completion) is introduced to combine the well-known systems performing based on popularity and neural language model. Furthermore, this system is supplemented by time-aware features that blend both context and semantic information in a collaborative manner. Experimental studies on the standard AOL dataset are conducted to compare our proposed system with state-of-the-art methods, that is, FactorCell, ConcatCell and Unadapted. The results illustrate the significant superiorities of TIPS in terms of mean reciprocal rank (MRR), especially for short-length prefixes.

Personalized Query Auto-Completion for Large-Scale POI Search at Baidu Maps

Query auto-completion (QAC) is a featured function that has been widely adopted by many sub-domains of search. It can dramatically reduce the number of typed characters and avoid spelling mistakes. These merits of QAC are highlighted to improve user satisfaction, especially when users intend to type in a query on mobile devices. In this article, we will present our industrial solution to the personalized QAC for the point of interest (POI) search at Baidu Maps, a well-known Web mapping service on mobiles in China. The industrial solution makes a good tradeoff between the offline effectiveness of a novel neural learning model that we devised for feature generation and the online efficiency of an off-the-shelf learning to rank (LTR) approach for the real-time suggestion. Besides some practical lessons from how a real-world QAC system is built and deployed in Baidu Maps to facilitate a large number of users in searching tens of millions of POIs, we mainly explore two specific features for the personalized QAC function of the POI search engine: the spatial-temporal characteristics of POIs and the historically queried POIs of individual users . We leverage the large-volume POI search logs in Baidu Maps to conduct offline evaluations of our personalized QAC model measured by multiple metrics, including Mean Reciprocal Rank (MRR), Success Rate (SR), and normalized Discounted Cumulative Gain (nDCG). Extensive experimental results demonstrate that the personalized model enhanced by the proposed features can achieve substantial improvements (i.e., +3.29% MRR, +3.78% SR@1, +5.17% SR@3, +1.96% SR@5, and +3.62% nDCG@5). After deploying this upgraded model into the POI search engine at Baidu Maps for A/B testing online, we observe that some other critical indicators, such as the average number of keystrokes and the average typing speed at keystrokes in a QAC session, which are also related to user satisfaction, decrease as well by 1.37% and 1.69%, respectively. So the conclusion is that the two kinds of features contributed by us are quite helpful in personalized mapping services for industrial practice.

GFocus: User Focus-based Graph Query Autocompletion

Graph query autocompletion ( <sc xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">gQAC</small> ) generates a small list of ranked query suggestions during the graph query formulation process in a visual environment. The current state-of-the-art of <sc xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">gQAC</small> provides suggestions that are formed by adding subgraph increments to arbitrary places of an existing (partial) user query. However, according to the research results on human-computer interaction (HCI), humans can only interact with a small number of recent software artifacts in hand. Hence, many of such suggestions could be irrelevant. In this paper, we present the <sc xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">GFocus</small> framework that exploits a novel notion of <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">user focus of graph query formulation</i> (or simply <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">focus</i> ). Intuitively, the focus is the subgraph that a user is working on. We formulate <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">locality principles</i> inspired by the HCI research to automatically identify and maintain the focus. We propose novel monotone submodular ranking functions for generating <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">popular</i> and <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">comprehensive</i> query suggestions only at the focus. In particular, the query suggestions of <sc xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">GFocus</small> have high result counts (when they are used as queries) and maximally cover the possible suggestions at the focus. We propose efficient algorithms and an index for ranking the suggestions. Our results show that <sc xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">GFocus</small> saves 12-32 percent more mouse clicks and is 35× more efficient than the state-of-the-art competitor.

Efficient query autocompletion with edit distance-based error tolerance

Query autocompletion is an important feature saving users many keystrokes from typing the entire query. In this paper, we study the problem of query autocompletion that tolerates errors in users’ input using edit distance constraints. Previous approaches index data strings in a trie, and continuously maintain all the prefixes of data strings whose edit distances from the query string are within the given threshold. The major inherent drawback of these approaches is that the number of such prefixes is huge for the first few characters of the query string and is exponential in the alphabet size. This results in slow query response even if the entire query approximately matches only few prefixes. We propose a novel neighborhood generation-based method to process error-tolerant query autocompletion. Our proposed method only maintains a small set of active nodes, thus saving both space and time to process the query. We also study efficient duplicate removal, a core problem in fetching query answers, and extend our method to support top-k queries. Optimization techniques are proposed to reduce the index size. The efficiency of our method is demonstrated through extensive experiments on real datasets.

Cohort-based personalized query auto-completion

Query auto-completion (QAC) facilitates query formulation by predicting completions for given query prefix inputs. Most web search engines use behavioral signals to customize query completion lists for users. To be effective, such personalized QAC models rely on the access to sufficient context about each user’s interest and intentions. Hence, they often suffer from data sparseness problems. For this reason, we propose the construction and application of cohorts to address context sparsity and to enhance QAC personalization. We build an individual’s interest profile by learning his/her topic preferences through topic models and then aggregate users who share similar profiles. As conventional topic models are unable to automatically learn cohorts, we propose two cohort topic models that handle topic modeling and cohort discovery in the same framework. We present four cohort-based personalized QAC models that employ four different cohort discovery strategies. Our proposals use cohorts’ contextual information together with query frequency to rank completions. We perform extensive experiments on the publicly available AOL query log and compare the ranking effectiveness with that of models that discard cohort contexts. Experimental results suggest that our cohort-based personalized QAC models can solve the sparseness problem and yield significant relevance improvement over competitive baselines.

Abstraction of query auto completion logs for anonymity-preserving analysis

Query auto completion (QAC) is used in search interfaces to interactively offer a list of suggestions to users as they enter queries. The suggested completions are updated each time the user modifies their partial query, as they either add further keystrokes or interact directly with completions that have been offered. In this work we use a state model to capture the possible interactions that can occur in a QAC environment. Using this model, we show how an abstract QAC log can be derived from a sequence of QAC interactions; this log does not contain the actual characters entered, but records only the sequence of types of interaction, thus preserving user anonymity with extremely high confidence. To validate the usefulness of the approach, we use a large scale abstract QAC log collected from a popular commercial search engine to demonstrate how previous and new knowledge about QAC behavior can be inferred without knowledge of the queries being entered. An interaction model is then derived from this log to demonstrate its validity, and we report observations on user behavior with QAC systems based on the interaction model that is proposed.

Modeling concepts and their relationships for corpus-based query auto-completion

AbstractQuery auto-completion helps users to formulate their information needs by providing suggestion lists at every typed key. This task is commonly addressed by exploiting query logs and the approaches proposed in the literature fit well in web scale scenarios, where usually huge amounts of past user queries can be analyzed to provide reliable suggestions. However, when query logs are not available, e.g. in enterprise or desktop search engines, these methods are not applicable at all. To face these challenging scenarios, we present a novel corpus-based approach which exploits the textual content of an indexed document collection in order to dynamically generate query completions. Our method extracts informative text fragments from the corpus and it combines them using a probabilistic graphical model in order to capture the relationships between the extracted concepts. Using this approach, it is possible to automatically complete partial queries with significant suggestions related to the keywords already entered by the user without requiring the analysis of the past queries. We evaluate our system through a user study on two different real-world document collections. The experiments show that our method is able to provide meaningful completions outperforming the state-of-the art approach.

Location-aware query reformulation for search engines

Query reformulation, including query recommendation and query auto-completion, is a popular add-on feature of search engines, which provide related and helpful reformulations of a keyword query. Due to the dropping prices of smartphones and the increasing coverage and bandwidth of mobile networks, a large percentage of search engine queries are issued from mobile devices. This makes it possible to improve the quality of query recommendation and auto-completion by considering the physical locations of the query issuers. However, limited research has been done on location-aware query reformulation for search engines. In this paper, we propose an effective spatial proximity measure between a query issuer and a query with a location distribution obtained from its clicked URLs in the query history. Based on this, we extend popular query recommendation and auto-completion approaches to our location-aware setting, which suggest query reformulations that are semantically relevant to the original query and give results that are spatially close to the query issuer. In addition, we extend the bookmark coloring algorithm for graph proximity search to support our proposed query recommendation approaches online, and we adapt an A* search algorithm to support our query auto-completion approach. We also propose a spatial partitioning based approximation that accelerates the computation of our proposed spatial proximity. We conduct experiments using a real query log, which show that our proposed approaches significantly outperform previous work in terms of quality, and they can be efficiently applied online.

Query Auto-Completion Based on Word2vec Semantic Similarity

Query auto-completion (QAC) is the first step of information retrieval, which helps users formulate the entire query after inputting only a few prefixes. Regarding the models of QAC, the traditional method ignores the contribution from the semantic relevance between queries. However, similar queries always express extremely similar search intention. In this paper, we propose a hybrid model FS-QAC based on query semantic similarity as well as the query frequency. We choose word2vec method to measure the semantic similarity between intended queries and pre-submitted queries. By combining both features, our experiments show that FS-QAC model improves the performance when predicting the user’s query intention and helping formulate the right query. Our experimental results show that the optimal hybrid model contributes to a 7.54% improvement in terms of MRR against a state-of-the-art baseline using the public AOL query logs.

String Similarity Search: A Hash-Based Approach

String similarity search is a fundamental query that has been widely used for DNA sequencing, error-tolerant query auto-completion, and data cleaning needed in database, data warehouse, and data mining. In this paper, we study string similarity search based on edit distance that is supported by many database management systems such as Oracle and PostgreSQL . Given the edit distance, ${\mathsf {ed}} (s,t)$ , between two strings, $s$ and $t$ , the string similarity search is to find every string $t$ in a string database $D$ which is similar to a query string $s$ such that ${\mathsf {ed}} (s, t) \leq \tau$ for a given threshold $\tau$ . In the literature, most existing work takes a filter-and-verify approach, where the filter step is introduced to reduce the high verification cost of two strings by utilizing an index built offline for $D$ . The two up-to-date approaches are prefix filtering and local filtering. In this paper, we study string similarity search where strings can be either short or long. Our approach can support long strings, which are not well supported by the existing approaches due to the size of the index built and the time to build such index. We propose two new hash-based labeling techniques, named $\mathsf {OX}$ label and $\mathsf {XX}$ label, for string similarity search. We assign a hash-label, ${\mathsf {H}} _s$ , to a string $s$ , and prune the dissimilar strings by comparing two hash-labels, ${\mathsf {H}} _s$ and ${\mathsf {H}} _t$ , for two strings $s$ and $t$ in the filter step. The key idea is to take the dissimilar bit-patterns between two hash-labels. We discuss our hash-based approaches, address their pruning power, and give the algorithms. Our hash-based approaches achieve high efficiency, and keep its index size and index construction time one order of magnitude smaller than the existing approaches in our experiment at the same time.

Learning Online Trends for Interactive Query Auto-Completion

Query auto-completion (QAC) is widely used by modern search engines to assist users by predicting their intended queries. Most QAC approaches rely on deterministic batch learning algorithms trained from past query log data. However, query popularities keep changing all the time and QAC operates in a real-time scenario where users interact with the search engine continually. So, ideally, QAC must be timely and adaptive enough to reflect time-sensitive changes in an online fashion. Second, due to the vertical position bias, a query suggestion with a higher rank tends to attract more clicks regardless of user’s original intention. Hence, in the long run, it is important to place some lower ranked yet potentially more relevant queries to higher positions to collect more valuable user feedbacks. In order to tackle these issues, we propose to formulate QAC as a ranked Multi-Armed Bandits (MAB) problem which enjoys theoretical soundness. To utilize prior knowledge from query logs, we propose to use Bayesian inference and Thompson Sampling to solve this MAB problem. Extensive experiments on large scale datasets show that our QAC algorithm has the capacity to adaptively learn temporal trends, and outperforms existing QAC algorithms in ranking qualities.

Interrogation Based on Semantic Annotations

Traditional information search approaches do not explicitly capture the meaning of a keyword query, but provide a good way for the user to express his or her information needs based on the keywords. In principle, semantic search aims to produce better results than traditional keyword search, but its progression has retarded because of to the complexity of the query languages. In this article, the authors present an approach to adapt keyword queries to querying the semantic web based on semantic annotations: the approach automatically construct structured formal queries from keywords. The authors propose a new process where they introduce a novel context-based query autocompletion feature to help the users to construct their keywords query by suggesting queries given prefixes. They also address the problem of context-based generating formal queries by exploiting user's query history, where previous queries can be used as contextual information for generating a new query. With the first tests, the authors' approach achieved encouraging results.

Term-level semantic similarity helps time-aware term popularity based query completion

Query completion service, normally known in the form of query auto completion (QAC) and widely provided by common search engines, assists users to formulate their queries after only typing few keystrokes. Previous work on QAC basically ranks query candidates according to their query popularity whic h is collected from the search logs, ignoring the internal semantic similarity between terms inside a query. However, we argue semantically related terms are apt to be combined when generating a query. In addition, as users often engage in QAC at word boundary (i.e., after typing a full word), we suppose that the time-aware popularity of the first word in a query candidate could affect the ranking of QAC candidates. Hence, based on the Markov assumption, we propose a new QAC ranking method, which models the QAC engagement as a Markov Chain and takes the semantic similarity between query terms into account. We contrast our proposed model with the traditional query popularity-based QAC approaches and verify its effectiveness in terms of Mean Reciprocal Rank (MRR). The experimental results show that our model significantly outperforms the baselines, achieving an average MRR improvement around 4% over the baselines.

The use of query auto-completion over the course of search sessions with multifaceted information needs

Query auto-completion (QAC) is the ubiquitous information search function that displays a list of suggested queries, where the list changes as the searcher types. This article reports on an exploratory study of QAC usage during complete search sessions in a lab study of 29 participants, where a session comprised searching on an assigned multi-faceted task. While prior research has reported average usage rates independent of the structure of search sessions, our findings are the first to describe when QAC was used within whole sessions and associations between usage and search task subtopics. Results show the value of QAC in shorter sessions and higher retrieval performance. Importantly, results also show that when QAC was used, it was most likely for the first query of a session, that use was less likely for subsequent queries, and that when the first query of a session did not use QAC, subsequent use was far less likely. The findings have implications for the value of QAC models that use knowledge from prior queries within a session to converge on optimal suggestions over successive queries. The findings are important for development of useful query assistance mechanisms for searchers. The study leads to new research questions on the effect of reformulation patterns on QAC usefulness and searchers’ attention to QAC throughout search sessions.