Traditional Inverted Index Research Articles

ColBERT-PRF: Semantic Pseudo-Relevance Feedback for Dense Passage and Document Retrieval

Pseudo-relevance feedback mechanisms, from Rocchio to the relevance models, have shown the usefulness of expanding and reweighting the users’ initial queries using information occurring in an initial set of retrieved documents, known as the pseudo-relevant set. Recently, dense retrieval – through the use of neural contextual language models such as BERT for analysing the documents’ and queries’ contents and computing their relevance scores – has shown a promising performance on several information retrieval tasks still relying on the traditional inverted index for identifying documents relevant to a query. Two different dense retrieval families have emerged: the use of single embedded representations for each passage and query, e.g., using BERT’s [CLS] token, or via multiple representations, e.g., using an embedding for each token of the query and document (exemplified by ColBERT). In this work, we conduct the first study into the potential for multiple representation dense retrieval to be enhanced using pseudo-relevance feedback and present our proposed approach ColBERT-PRF. In particular, based on the pseudo-relevant set of documents identified using a first-pass dense retrieval, ColBERT-PRF extracts the representative feedback embeddings from the document embeddings of the pseudo-relevant set. Among the representative feedback embeddings, the embeddings that most highly discriminate among documents are employed as the expansion embeddings, which are then added to the original query representation. We show that these additional expansion embeddings both enhance the effectiveness of a reranking of the initial query results as well as an additional dense retrieval operation. Indeed, experiments on the MSMARCO passage ranking dataset show that MAP can be improved by up to 26% on the TREC 2019 query set and 10% on the TREC 2020 query set by the application of our proposed ColBERT-PRF method on a ColBERT dense retrieval approach.We further validate the effectiveness of our proposed pseudo-relevance feedback technique for a dense retrieval model on MSMARCO document ranking and TREC Robust04 document ranking tasks. For instance, ColBERT-PRF exhibits up to 21% and 14% improvement in MAP over the ColBERT E2E model on the MSMARCO document ranking TREC 2019 and TREC 2020 query sets, respectively. Additionally, we study the effectiveness of variants of the ColBERT-PRF model with different weighting methods. Finally, we show that ColBERT-PRF can be made more efficient, attaining up to 4.54× speedup over the default ColBERT-PRF model, and with little impact on effectiveness, through the application of approximate scoring and different clustering methods.

A Keyword-Grouping Inverted Index Based Multi-Keyword Ranked Search Scheme Over Encrypted Cloud Data

With the comprehensive development of cloud computing technology, more and more enterprises and individuals tend to outsource computing, data, and other resources to the cloud service providers to save the data management cost. Since the plaintext data outsourcing in the cloud could leak users’ private information, it is highly recommended to encrypt them before outsourcing. However, it is a challenge to perform searches over encrypted cloud data. In this paper, we adopt the keyword grouping idea into the traditional inverted index and propose a keyword-grouping inverted index (KGI-index). Based on the index, we propose a privacy-preserving KGI-index based multi-keywords ranked search scheme (KMRS). To improve the search efficiency, we adopt two strategies including grouping high relevant keywords and using the complete binary tree structure to optimize the index. The security analysis and experimental result show that the proposed scheme is a privacy-preserving and efficient multi-keyword ranked search scheme over encrypted cloud data.

Using Inverted Index for Fingerprint Search

Fingerprints are the most used biometric information for identifying people. With the increase in fingerprint data, indexing techniques are essential to perform an efficient search. In this work, we devise a solution that applies traditional inverted index, widely used in textual information retrieval, for fingerprint search. For that, it first converts fingerprints to text documents using techniques, such as Minutia Cylinder-Code and Locality-Sensitive Hashing, and then indexes them in inverted files. In the experimental evaluation, our approach obtained 0.42% of error rate with 10% of penetration rate in the FVC2002 DB1a data set, surpassing some established methods.

A verifiable ranked ciphertext retrieval scheme based on bilinear mapping

SummaryIt is prevalent nowadays for data owners to outsource their data to the cloud. Most of the traditional searchable schemes are proposed under the honest and curious model, lacking verification of integrity and correctness of the retrieval results. Because the cloud server is not completely trustworthy and there may be improper execution of the users' retrieval requests, it is necessary to verify the retrieval results returned by the cloud server. In order to address these issues, we propose a verifiable ranked ciphertext retrieval scheme (VRCRS) based on bilinear mapping. For the purpose of ranking the keywords and improving the security of ciphertext retrieval scheme, we utilize the lucene search engine toolkit to score keywords, and improve the traditional inverted index structure for building a secure inverted index structure. By adding dummy words into the dictionary of keywords, the proposed scheme can resist the keywords statistics attack of malicious server. To ensure the correctness and integrity of returned ciphertext, we apply bilinear mapping to generate validation tags for keywords. The theoretical analysis and experimental results show that the proposed scheme is secure and efficient. VRCRS can actually identify the illegal behavior of cloud server such as tampering and forgery.

Pavo: A RNN-Based Learned Inverted Index, Supervised or Unsupervised?

The booms of big data and graphic processing unit technologies have allowed us to explore more appropriate data structures and algorithms with smaller time complexity. However, the application of machine learning as a potential alternative for the traditional data structures, especially using deep learning, is a relatively new and largely uncharted territory. In this paper, we propose a novel recurrent neural network-based learned inverted index, called Pavo, to efficiently and flexibly organize inverted data. The basic hash function in the traditional inverted index is replaced by a hierarchical neural network, which makes Pavo be able to well adapt for various data distributions while showing lower collision rate as well as higher space utilization rate. A particular feature of our approach is that a novel unsupervised learning strategy to construct the hash function is proposed. To the best of our knowledge, there are no similar results, in which the unsupervised learning strategy is employed to design hash functions, in the existing literature. Extensive experimental results show that the unsupervised model owns some advantages than the supervised one. Our approaches not only demonstrate the feasibility of deep learning-based data structures for index purpose but also provide benefits for developers to make more accurate decisions on both the design and the configuration of data organization, operation, and parameters tuning of neural network so as to improve the performance of information searching.

CCQA: A Chinese Community Question-answering Tool Based on Sentence Similarity with Word Embedding

Online CQA (community question-answering) systems are a hot topic today. Users can freely ask and answer questions in any subject with little restrictions. Such communities are a good resource for automatic question-answering systems, providing resourceful natural-language question-answer pairs with sub-attributes like rankings, scores, likes and so forth for evaluation. Traditional question retrieval engines use the same word for searching but neglect similar expressions, which may lead to lexical gap, omitting results with similar expressions. In this article, we build a question-answering system in Chinese based on corpuses selected from online Chinese cQAs by selecting the most similar query in the database with trained word embedding. Results show our method performs better in bridging lexical gap than traditional inverted index.

TEII: Topic enhanced inverted index for top-k document retrieval

In recent years, topic modeling is gaining significant momentum in information retrieval (IR). Researchers have found that utilizing the topic information generated through topic modeling together with traditional TF-IDF information generates superior results in document retrieval. However, in order to apply this idea to real-life IR systems, some critical problems need to be solved: how to store the topic information and how to utilize it with the TF-IDF information for efficient document retrieval. In this paper, we propose the Topic Enhanced Inverted Index (TEII) to incorporate the topic information into the inverted index for efficient top-k document retrieval. Specifically, we explore two different types of TEIIs. We first propose the incremental TEII, which includes the topic information into the traditional inverted index by adding topic-based inverted lists. The incremental TEII is beneficial for legacy IR systems, since it does not change the existing TF-IDF-based inverted lists. As a more flexible alternative, we propose the hybrid TEII to incorporate the topic information into each posting of the inverted index. In the hybrid TEII, two relaxation methods are proposed to support dynamic estimation of the upper bound impact of each posting. The hybrid TEII is highly extensible for incorporating different ranking factors and we show an extension of the hybrid TEII by considering the static quality of the documents in the corpus. Based on the incremental and hybrid TEIIs, we develop several query processing algorithms to support efficient top-k document retrieval on TEIIs. Empirical evaluation on the TREC dataset verifies the effectiveness and efficiency of the proposed index structures and query processing algorithms.

Ginix: Generalized inverted index for keyword search

Keyword search has become a ubiquitous method for users to access text data in the face of information explosion. Inverted lists are usually used to index underlying documents to retrieve documents according to a set of keywords efficiently. Since inverted lists are usually large, many compression techniques have been proposed to reduce the storage space and disk I/O time. However, these techniques usually perform decompression operations on the fly, which increases the CPU time. This paper presents a more efficient index structure, the Generalized INverted IndeX (Ginix), which merges consecutive IDs in inverted lists into intervals to save storage space. With this index structure, more efficient algorithms can be devised to perform basic keyword search operations, i.e., the union and the intersection operations, by taking the advantage of intervals. Specifically, these algorithms do not require conversions from interval lists back to ID lists. As a result, keyword search using Ginix can be more efficient than those using traditional inverted indices. The performance of Ginix is also improved by reordering the documents in datasets using two scalable algorithms. Experiments on the performance and scalability of Ginix on real datasets show that Ginix not only requires less storage space, but also improves the keyword search performance, compared with traditional inverted indexes.

A social inverted index for social-tagging-based information retrieval

Keywords have played an important role not only for searchers who formulate a query, but also for search engines that index documents and evaluate the query. Recently, tags chosen by users to annotate web resources are gaining significance for improving information retrieval (IR) tasks, in that they can act as meaningful keywords bridging the gap between humans and machines. One critical aspect of tagging (besides the tag and the resource) is the user (or tagger); there exists a ternary relationship among the tag, resource, and user. The traditional inverted index, however, does not consider the user aspect, and is based on the binary relationship between term and document. In this paper we propose a social inverted index – a novel inverted index extended for social-tagging-based IR – that maintains a separate user sublist for each resource in a resource-posting list to contain each user’s various features as weights. The social inverted index is different from the normal inverted index in that it regards each user as a unique person, rather than simply count the number of users, and highlights the value of a user who has participated in tagging. This extended structure facilitates the use of dynamic resource weights, which are expected to be more meaningful than simple user-frequency-based weights. It also allows a flexible response to the conditional queries that are increasingly required in tag-based IR. Our experiments have shown that this user-considering indexing performs better in IR tasks than a normal inverted index with no user sublists. The time and space overhead required for index construction and maintenance was also acceptable.

A Design of the Inverted Index Based on Web Document Comprehending

A design of inverted index based on web document comprehending is induced and the algorithm is presented by combining the inverted index and the web documents understanding technology. Experimental results show that less time is cost than the traditional inverted index to query in the documents with the same size.

An effective 3-in-1 keyword search method over heterogeneous data sources

Conventional keyword search engines are restricted to a given data model and cannot easily adapt to unstructured, semi-structured or structured data. In this paper, we propose an efficient and adaptive keyword search method, called EASE, for indexing and querying large collections of heterogeneous data. To achieve high efficiency in processing keyword queries, we first model unstructured, semi-structured and structured data as graphs, and then summarize the graphs and construct graph indices instead of using traditional inverted indices. We propose an extended inverted index to facilitate keyword-based search, and present a novel ranking mechanism for enhancing search effectiveness. We have conducted an extensive experimental study using real datasets, and the results show that EASE achieves both high search efficiency and high accuracy, and outperforms the existing approaches significantly.

An effective and versatile keyword search engine on heterogenous data sources

We present EASE, an effective and versatile keyword search engine that enables users to easily access the heterogenous data composed of unstructured, semi-structured and structured data, without the need of learning XPath/XQuery or SQL languages. EASE addresses a challenge in keyword search that has been neglected in the literature: how to efficiently and adaptively process keyword queries on the heterogenous data. To provide such capability, EASE models unstructured, semi-structured and structured data as graphs, summarizes the graphs, and constructs graph indices instead of using traditional inverted indices for effective keyword search. EASE adopts an extended inverted index to facilitate keyword-based search, and employs a novel ranking mechanism for enhancing search effectiveness.