Document Ranking Research Articles

Relation-aware deep neural network enables more efficient biomedical knowledge acquisition from massive literature

Biomedical knowledge is typically organized in a relational scheme, such as chemical-disease relation, gene-disease relation, and gene-pathway relation. Biomedical scientists heavily rely on search engines to acquire up-to-date relational knowledge from massive biomedical articles. The navigation efficiency of the retrieval process, however, is significantly restricted by keyword matching techniques unaware of the biomedical relations of these keywords in articles. To bridge the gap between existing retrieval techniques and practical access demands for relational knowledge, we present a novel framework, Biomedical Relation-Aware Document Ranking (BioRADR), capable of retrieving articles expressing specific relations with respect to the queried entity pair. Based on a deep neural network, BioRADR can be trained from large-scale data automatically annotated via distant supervision, and empirical evaluation reveals that it outperforms the strongest baseline by over 8 points in NDCG@1. We implement an online system (http://bioradr.ai.thunlp.org/) based on BioRADR, enabling more efficient relation-oriented retrieval of biomedical articles.

An efficient long-text semantic retrieval approach via utilizing presentation learning on short-text

Although the short-text retrieval model by BERT achieves significant performance improvement, research on the efficiency and performance of long-text retrieval still faces challenges. Therefore, this study proposes an efficient long-text retrieval model based on BERT (called LTR-BERT). This model achieves speed improvement while retaining most of the long-text retrieval performance. In particular, The LTR-BERT model is trained by using the relevance between short texts. Then, the long text is segmented and stored off-line. In the retrieval stage, only the coding of the query and the matching scores are calculated, which speeds up the retrieval. Moreover, a query expansion strategy is designed to enhance the representation of the original query and reserve the encoding region for the query. It is beneficial for learning missing information in the representation stage. The interaction mechanism without training parameters takes into account the local semantic details and the whole relevance to ensure the accuracy of retrieval and further shorten the response time. Experiments are carried out on MS MARCO Document Ranking dataset, which is specially designed for long-text retrieval. Compared with the interaction-focused semantic matching method by BERT-CLS, the MRR@10 values of the proposed LTR-BERT method are increased by 2.74%. Moreover, the number of documents processed per millisecond increased by 333 times.

A Detailed Comparative Analysis of Document Ranking Approaches

A Detailed Comparative Analysis of Document Ranking Approaches

Integrating Representation and Interaction for Context-Aware Document Ranking

Recent studies show that historical behaviors (such as queries and their clicks) contained in a search session can benefit the ranking performance of subsequent queries in the session. Existing neural context-aware ranking models usually rank documents based on either latent representations of user search behaviors or the word-level interactions between the candidate document and each historical behavior in the search session. However, these two kinds of models both have their own drawbacks. Representation-based models neglect fine-grained information on word-level interactions, whereas interaction-based models suffer from the length restriction of session sequence because of the large cost of word-level interactions. To complement the limitations of these two kinds of models, we propose a unified context-aware document ranking model that takes full advantage of both representation and interaction. Specifically, instead of matching a candidate document with every single historical query in a session, we encode the session history into a latent representation and use this representation to enhance the current query and the candidate document. We then just match the enhanced query and candidate document with several matching components to capture the fine-grained information of word-level interactions. Rich experiments on two public query logs prove the effectiveness and efficiency of our model for leveraging representation and interaction.

Named Entity Recognition based Resume Parser and Summarizer

Resumes are unstructured documents and the language used in the resume will be ambiguous and variable. Screening documents in large amounts is a difficult operation, and recruiters and hiring managers waste a lot of time going through each and every CV of the candidates. The extraction of valuable information from a resume is difficult due to this variability. It necessitates an urgent need to comprehend the context in which words are used. Using spaCy, a powerful natural language processing (NLP) package, this study suggests a machine learning method to phrase matching in resumes, focused on the extraction of specific entities. Like a human recruiter, it can examine and extract comprehensive information from resumes. While parsing, it maintains track of the terms in order to classify people. Specific Important entities are extracted from the resume document and then it is stored for later classification. Document ranking is processed and the final rankings are depicted in descending order with respect to the score of the resume in order to make it easy for the recruiters to shortlist the resumes quickly and efficiently. Recruiters may pick the needed applicants based on the scores rather than sifting through stacks of resumes from unqualified prospects.

A fuzzy content recommendation system using similarity analysis, content ranking and clustering

Recently, the e-learners are drastically increased from the last two decades. Everything is learnt through internet without help of the tutor as well. For this purpose, the e-learners are required more e-learning applications that are able to supply optimal and satisfied data based on their capability. No content recommendation system is available for recommending suitable contents to the learners. For this purpose, this paper proposes a new semantic and fuzzy aware content recommendation system for retrieving the suitable content for the users. In this content recommendation system, we propose two content pre-processing algorithms namely Target Keyword based Data Pre-processing Algorithm (TKDPA) and Intelligent Anova-T Residual Algorithm (IAATRA) for selecting the more relevant features from the document. Moreover, a new Fuzzy rule based Similarity Matching algorithm (FRSMA) is proposed and used in this system for finding the similarity between the two terms and also rank them by using the newly proposed Similarity and Temporal aware Weighted Document Ranking Algorithm (STWDRA). In addition, a content clustering process is also incorporated for gathering relevant content. Finally, a new Fuzzy, Target Keyword and Similarity Score based Content Recommendation Algorithm (FTKSCRA) is also proposed for recommending the more relevant content to the learners accurately. The experiments have been conducted for evaluating the proposed content recommendation system and proved as better than the existing recommendation systems in terms of precision, recall, f-measure and prediction accuracy.

Parallel Traversal of Large Ensembles of Decision Trees

Machine-learnt models based on additive ensembles of regression trees are currently deemed the best solution to address complex classification, regression, and ranking tasks. The deployment of such models is computationally demanding: to compute the final prediction, the whole ensemble must be traversed by accumulating the contributions of all its trees. In particular, traversal cost impacts applications where the number of candidate items is large, the time budget available to apply the learnt model to them is limited, and the users’ expectations in terms of quality-of-service is high. Document ranking in web search, where sub-optimal ranking models are deployed to find a proper trade-off between efficiency and effectiveness of query answering, is probably the most typical example of this challenging issue. This paper investigates multi/many-core parallelization strategies for speeding up the traversal of large ensembles of regression trees thus obtaining machine-learnt models that are, at the same time, effective, fast, and scalable. Our best results are obtained by the GPU-based parallelization of the state-of-the-art algorithm, with speedups of up to 102.6x.

A Survey of Designing Multilingual Document Retrieval and Ranking in Cloud

A Survey of Designing Multilingual Document Retrieval and Ranking in Cloud

Quantum contextuality in classical information retrieval

Document ranking based on probabilistic evaluations of relevance is known to exhibit non-classical correlations, which may be explained by admitting a complex structure of the event space, namely, by assuming the events to emerge from multiple sample spaces. The structure of event space formed by overlapping sample spaces is known in quantum mechanics, they may exhibit some counter-intuitive features, called quantum contextuality. In this Note I observe that from the structural point of view quantum contextuality looks similar to personalization of information retrieval scenarios. Along these lines, Knowledge Revision is treated as operationalistic measurement and a way to quantify the rate of personalization of Information Retrieval scenarios is suggested.

An Efficient Document Ranking Algorithm and Classification for Recommending E-Content

E-learning is an important education media in the current internet scenario. However, selection of suitable content for effective and fast learning is a challenging problem to the learners. E-learning content selection through group discussions using social networks is a new approach that helps the learners to choose the suitable content from the web. For this purpose, we propose a new content recommendation system for recommending suitable e-content for the e-learners. This content recommendation system uses an existing classification algorithm and the proposed ranking algorithm for making effective decision over the e-content for the respective users.

Research on Personalized Document Retrieval and Ranking Strategy

In order to obtain proper text information consistent with the users’ cognitive level, a new personalized retrieval method is proposed for the distant learners. At first the user’s personalized cognitive structure is established and this structure can represent the users’ cognitive level formally. Then related concept can be extended around the keywords, which usually reflect their focus of attention. How to extend depennd on the semantic distance which should be measured by some new formula in this article. At last these queried documents should be ranked according to the user’s cognitive level and the difficulty of the document containt. Experiments show that even if some different learners use the same keywords, the query result is consistent with their cognitive level respectively.

A Novel Approach for Document Ranking in Digital Libraries using Extractive Summarization

Digital libraries have become an important tool for searching the scientific literature. The growth of digital libraries is exponential based upon the fact that researchers are utilizing it more and more by the days. To make the search better, the content of digital libraries should be ranked properly. The rank of research paper depends upon many factors like citations to paper; content, authors and publications of the paper etc. Based upon these parameters, different ranking algorithms have been proposed till now. In this paper, a new algorithm C3 ranking is proposed, which considers two important factors i.e. citations to the paper and the relevancy of the content with the query.

Document ranking with quantum probabilities

In this thesis we investigate the use of quantum probability theory for ranking documents. Quantum probability theory is used to estimate the probability of relevance of a document given a user's query. We posit that quantum probability theory can lead to a better estimation of the probability of a document being relevant to a user's query than the common IR approach, i. e. the Probability Ranking Principle (PRP), which is based upon Kolmogorovian probability theory. Following our hypothesis, we formulate an analogy between the document retrieval scenario and a physical scenario, that of the double slit experiment. Through the analogy, we propose a novel ranking approach, the quantum probability ranking principle (qPRP). Key to our proposal is the presence of quantum interference. Mathematically, this is the statistical deviation between empirical observations and expected values predicted by the Kolmogorovian rule of additivity of probabilities of disjoint events in configurations such that of the double slit experiment. While PRP explicitly assumes that the relevancy of a document is independent of that of other documents, we suggest that qPRP implicitly models interdependent document relevance through quantum interference and thus is suited to those document ranking tasks where the independence assumption fails. Throughout the thesis, we also suggest how quantum interference can be estimated for effective document ranking. To validate our proposal and to gain more insights about approaches for document ranking, we (1) analyse PRP, qPRP and other ranking approaches, exposing the assumptions underlying their ranking criteria and formulating the conditions for the optimality of the two ranking principles, (2) empirically compare three ranking principles (i. e. PRP, interactive PRP, and qPRP) and two state-of-the-art ranking strategies in two retrieval scenarios, those of ad-hoc retrieval and diversity retrieval, (3) analytically contrast the ranking criteria of the examined approaches, exposing similarities and differences, (4) study the ranking behaviours of approaches alternative to PRP in terms of the kinematics they impose on relevant documents, i. e. by considering the extent and direction of the movements of relevant documents across the ranking recorded when comparing PRP against its alternatives. Our findings show that the effectiveness of the examined ranking approaches strongly depends upon the evaluation context. In the traditional evaluation context of ad-hoc retrieval, PRP is empirically shown to be better than or comparable to alternative ranking approaches. However, when evaluation contexts that account for interdependent document relevance are examined (i. e. when the relevance of a document is assessed also with respect to other retrieved documents, as it is the case in the diversity retrieval scenario), the use of quantum probability theory and thus of qPRP is shown to improve retrieval and ranking effectiveness over the traditional PRP and alternative ranking strategies, such as Maximal Marginal Relevance, Portfolio theory, and Interactive PRP. This work represents a significant step forward regarding the use of quantum theory in information retrieval. It demonstrates that the application of quantum theory to problems within information retrieval can lead to improvements both in modelling power and retrieval effectiveness, allowing the constructions of models that capture the complexity of information retrieval situations. Furthermore, the thesis opens up a number of lines of future research. These include investigating estimations and approximations of quantum interference in qPRP, exploiting complex numbers for the representation of documents and queries, and applying the concepts underlying qPRP to tasks other than document ranking. This dissertation was completed at School of Computing Science, University of Glasgow under the advise of Dr. Leif Azzopardi and Prof. Keith van Rijsbergen. Prof. Norbert Fuhr, Dr. Iadh Ounis, and Dr. John O'Donnell served as dissertation committee members. For the full dissertation, visit: http://theses.gla.ac.uk/3463.

A Novel Document Ranking Algorithm That Supports Mobile Healthcare Information Access Effectiveness

This study presented DROPT; an acronym for Document ranking Optmization algorithm approach, a new idea for the effectiveness of meaningful retrieval results from the information source. Proposed method extracted the frequency of query keyword terms that appears within the user context of Frequently Asked Questions (FAQ) systems on HIV/AIDS content related­ documents. The SMS messages were analyzed and then classified, with the aim of constructing a corpus of SMS related to HIV/AIDS. This study presented a novel framework of Information Retrieval Systems (IRS) based on the proposed algorithm. The developed DROPT procedure was used as an evaluation measure. This Term Frequency-Inverse Document Frequency (TFIDF) method was applied to obtain the experimental result that was found promising in ranking documents not only the order in which the relevant documents were retrieved, but also both the terms of the relevant documents in feedback and the terms of the irrelevant documents in feedback might be useful for relevance feedback, especially to define its fitness function (mean weight).

Incorporating rich features to boost information retrieval performance: A SVM-regression based re-ranking approach

Document ranking is an essential problem in the field of information retrieval (IR). Traditional weighting models such as BM25 and Language model can only take advantage of query terms. IR is a complex process that may be affected by a series of heterogeneous features. It is necessary to refine first-pass retrieval results by taking rich features into account. Traditional heuristic re-ranking approaches can only take advantage of a small number of homogeneous features that may affect information retrieval performance. In this paper, we propose and evaluate a regression-based document re-ranking approach for IR, in which we use SVM regression model to learn a re-ranking function automatically. Under this regression-based framework, we can take advantage of rich features to re-rank the firs-pass retrieved documents by traditional weighting models. We conduct a series of experiments on four standard IR collections in two different languages. The experimental results show that our proposed approach can significantly improve the retrieval performance over the first-pass retrieval. Moreover, by refining the first-pass retrieved document set, the traditional pseudo relevant feedback approaches can also be enhanced.

Progress towards an unassisted element identification from Laser Induced Breakdown Spectra with automatic ranking techniques inspired by text retrieval

In this communication, we will illustrate an algorithm for automatic element identification in LIBS spectra which takes inspiration from the vector space model applied to text retrieval techniques. The vector space model prescribes that text documents and text queries are represented as vectors of weighted terms (words). Document ranking, with respect to relevance to a query, is obtained by comparing the vectors representing the documents with the vector representing the query. In our case, we represent elements and samples as vectors of weighted peaks, obtained from their spectra. The likelihood of the presence of an element in a sample is computed by comparing the corresponding vectors of weighted peaks. The weight of a peak is proportional to its intensity and to the inverse of the number of peaks, in the database, in its wavelength neighboring. We suppose to have a database containing the peaks of all elements we want to recognize, where each peak is represented by a wavelength and it is associated with its expected relative intensity and the corresponding element. Detection of elements in a sample is obtained by ranking the elements according to the distance of the associated vectors from the vector representing the sample. The application of this approach to elements identification using LIBS spectra obtained from several kinds of metallic alloys will be also illustrated. The possible extension of this technique towards an algorithm for fully automated LIBS analysis will be discussed.

Ontology based Similarity Measure in Document Ranking

This paper presents a methodology for the ontology based semantic annotation of web pages with annotation weighting scheme that takes advantage of the different relevance of structured document fields. The retrieval model is based on the importance factors of the structural elements, which are used to re-rank the documents retrieval by the ontology based distance measure. The relevance concept similarity are combined with the annotation-weighting scheme to improve the relevance measures. The proposed method has been evaluated on USGS Science directory collection. Preliminary experiments results show that our method may generate relevant document in the top rank.

Searching structured documents

Structured document interchange formats such as XML and SGML are ubiquitous, however, information retrieval systems supporting structured searching are not. Structured searching can result in increased precision. A search for the author “Smith” in an unstructured corpus of documents specializing in iron-working could have a lower precision than a structured search for “Smith as author” in the same corpus. Analysis of XML retrieval languages identifies additional functionality that must be supported including searching at, and broken across multiple nodes in the document tree. A data structure is developed to support structured document searching. Application of this structure to information retrieval is then demonstrated. Document ranking is examined and adapted specifically for structured searching.

Document ranking based upon Markov chains

One of the most important problems in information retrieval is determining the order of documents in the answer returned to the user. Many methods and algorithms for document ordering have been proposed. The method introduced in this paper differs from them especially in that it uses a probabilistic model of a document set. In this model documents are regarded as states of a Markov chain, where transition probabilities are directly proportional to similarities between documents. Steady-state probabilities reflect similarities of particular documents to the whole answer set. If documents are ordered according to these probabilities, at the top of a list there will be documents that are the best representatives of the set, and at the bottom those which are the worst representatives. The method was tested against databases INSPEC and Networked Computer Science Technical Reference Library (NCSTRL). Test results are positive. Values of the Kendall rank correlation coefficient indicate high similarity between rankings generated by the proposed method and rankings produced by experts. Results are comparable with rankings generated by the vector model using standard weighting schema tf·idf.

Applying genetic algorithms to query optimization in document retrieval

This paper proposes a novel approach to automatically retrieve keywords and then uses genetic algorithms to adapt the keyword weights. One of the contributions of the paper is to combine the Bigram (Chen, A., He, J., Xu, L., Gey, F. C., & Meggs, J. 1997. Chinese text retrieval without using a dictionary, ACM SIGIR’97, Philadelphia, PA, USA, pp. 42–49; Yang, Y.-Y., Chang, J.-S., & Chen, K.-J. 1993), Document automatic classification and ranking, Master thesis, Department of Computer Science, National Tsing Hua University) model and PAT-tree structure (Chien, L.-F., Huang, T.-I., & Chien, M.-C. 1997 Pat-tree-based keyword extraction for Chinese information retrieval, ACM SIGIR’97, Philadelphia, PA, US, pp. 50–59) to retrieve keywords. The approach extracts bigrams from documents and uses the bigrams to construct a PAT-tree to retrieve keywords. The proposed approach can retrieve any type of keywords such as technical keywords and a person’s name. Effectiveness of the proposed approach is demonstrated by comparing how effective are the keywords found by both this approach and the PAT-tree based approach. This comparison reveals that our keyword retrieval approach is as accurate as the PAT-tree based approach, yet our approach is faster and uses less memory. The study then applies genetic algorithms to tune the weight of retrieved keywords. Moreover, several documents obtained from web sites are tested and experimental results are compared with those of other approaches, indicating that the proposed approach is highly promising for applications.