Pith. sign in

REVIEW 4 major objections 7 minor 64 references

Ontology-driven personalized information retrieval for XML documents

T0 review · 4 major / 7 minor · reviewed 2026-08-02 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read Adding ontology weighting and user-profile reinforcement to XML retrieval lifts precision from 0.426 to 0.710 and recall from 0.756 to 0.978 on the paper's test collection.

desk verdict Coherent but incremental framework; the evaluation is circular, so the reported precision/recall gains don't support the central claim. read the letter →

arxiv 2603.21139 v2 pith:77OZXZE5 submitted 2026-03-22 cs.IR cs.LG

Ounnaci Iddir , Ahmed-ouamer Rachid , Tai Dinh This is my paper
classification cs.IRcs.LG
keywords XMLinformationretrievalpersonalizedsearchdomainontologyuserprofileconceptweightingvectorspacemodelsemanticsimilarityprecisionandrecall
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper tries to establish that XML search improves when matching is driven by a domain ontology and a per-user profile rather than by keywords alone. It represents document fragments, queries, and user interests as vectors over ontology concepts, weights deeper concepts more heavily because they are more specific, and updates the profile after each query. On a 400-document computer-science corpus with 300 synthetic queries, it reports average precision rising from 0.426 to 0.710 and recall from 0.756 to 0.978 compared with a non-weighted keyword baseline. If true, the result would matter because the same query could return different, more targeted fragments for different users, and repeated searches would sharpen results.

What carries the argument

The load-bearing object is the ontology-weighted concept vector: document nodes, queries, and user profiles are all points in the same |C_Ω|-dimensional concept space. The three identities that carry the argument are (3), which redistributes the total ontology weight (normalized to 1) across concepts so deeper, more specific concepts weigh more; (6), which builds an element node's vector from descendant text nodes with a distance penalty and a coverage ratio, then scales by the user's interest in each concept; and (5), which updates profile weights after each query by adding exp(w_tj)−1 to the previous weight. Around these sit a cosine-similarity relevance score and an interval-encoding of t

What would settle it

Take the proposed weighted/profile-aware configuration and the keyword baseline to an independently annotated XML collection where human reviewers, not the ontology, decide relevance; compare precision, recall, and F1. The central claim fails if the proposed configuration no longer beats the baseline.

Watch

Extended reading notes

Core claim

The paper's central claim is that relevance in XML retrieval can be computed entirely in an ontology-concept space, and that doing so beats keyword matching. The system assigns each ontology concept a weight that rises with depth in the hierarchy, via W_R(C_k)=W_AVG+Δ(Coef(C_k)−Coef_AVG), so that leaf concepts describing precisely what a fragment is about dominate. Text nodes get concept weights from term frequency, an element-level inverse frequency, and the ontology weight; these weights propagate up the XML tree with a distance discount and a coverage factor, then are multiplied by the user's current interest weight. Each query also becomes a concept vector, and a fragment's score is the

Load-bearing premise

The experiments rely on a test collection whose documents, queries, and relevance judgments were all produced from the same domain ontology the system uses—if that ontology is not an independent standard of what is relevant, the reported gains may measure self-consistency rather than retrieval quality.

Editorial extensions

If this is right

  • XML search can return the relevant section (a text node or element subtree) instead of whole documents, because every node has its own concept vector and score.
  • Repeated queries by the same user should produce progressively better results, since each interaction reinforces the profile concepts of interest.
  • The same semantic vector machinery can be applied to any XML collection that can be mapped to a domain ontology, not just the computer-science corpus tested here.
  • Weighting specificity into concept weights means generic fragments are suppressed naturally, so precision gains should persist across topics within the ontology.
  • Because queries, documents, and profiles share one vector space, new users can be handled with a cold-start profile of uniform weights and then personalized once they search.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • A testable extension the paper leaves implicit: the profile update is unbounded exponential growth; normalizing or saturating the profile weights would test whether the reported stability across instants T1–T8 depends on this growth or survives a bounded variant.
  • The gains reported may largely reflect depth weighting rather than personalization; an ablation that adds ontology weights without the profile, and the profile without the weights, would separate the two contributions—Table 2's baseline removes both at once.
  • If corroborated on independent human-judged data, the same concept-vector approach could be carried to other semi-structured formats—annotated HTML, JSON with schemas—by defining a lightweight ontology mapping, though the paper argues XML is the natural fit.
  • The synthetic queries are generated from ontology concepts and keywords, so the evaluation measures how well the system re-finds documents organized by the same ontology; a real user's natural-language query may not map to concepts as cleanly as the generated ones.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 7 minor

Summary. The paper proposes a personalized XML information retrieval framework that combines a hand-built computer-science domain ontology with dynamically updated user profiles. Documents, queries, and profiles are represented as vectors of ontology concepts; deeper concepts receive larger weights (Eq. 3), text-node weights combine frequency with ontology weight (Eq. 4), element nodes are built by upward propagation (Eq. 6), and node-query matching uses cosine similarity together with a pertinence factor (Eq. 7). The user profile is updated with an exponential reinforcement rule (Eq. 5). The authors report precision, recall, and F1 results on a self-constructed 400-document XML collection with 300 ontology-derived queries, comparing against their own 2014 system rather than a keyword-based system.

Significance. If the empirical claim were established, the paper would offer a reasonably complete integration of ontology weighting, structural XML fragment retrieval, and adaptive user profiling. The formal notation is mostly clear, and the use of interval-based structural filtering and a detailed related-work taxonomy are useful. However, the central claim is currently unsupported: the corpus, queries, and relevance counts are all derived from the same ontology used for indexing, weighting, and profile building, so Tables 2 and 3 measure self-consistency rather than retrieval effectiveness. No external benchmark, independent relevance judgments, genuine keyword baseline, released code, or dataset is provided. The significance of the contribution therefore cannot be assessed from the present manuscript.

major comments (4)
  1. [Section 5.1, Tables 2-3] The central evaluation is circular. Section 5.1 states that the 400-document test base is 'organized according to the concepts of our domain ontology' and that the 300 queries were generated 'by selecting ontology concepts and using their keywords.' No relevance-assessment protocol is described. Therefore the 'relevant docs' counts in Table 2 are defined by the same ontology that Section 3.1 uses for concept weighting (Eq. 3), Section 4.1 uses for indexing (Eqs. 4 and 6), and Section 4.1.1 uses for profile construction (Eq. 5). The retrieval system and the ground truth share the same concept vocabulary by construction, so high precision/recall may reflect self-consistency rather than effectiveness against human information needs. An independent judgment pool or a reused collection with externally defined relevance is required; the dismissal of INEX/TREC/CLEF in Section 5.1 does not remov
  2. [Section 5.1, Tables 2-3] The comparison does not test the abstract's claim about 'keyword-based approaches.' The baseline is Ounnaci et al. (2014), the authors' earlier ontology-based system without concept weights and without the interest vector, not a keyword/BM25 or plain-text retrieval run. No such run is reported. Moreover, no confidence intervals, significance tests, or repeated-run variability are provided; Table 2 is based on five requests, and the precision/recall differences in Table 3 are presented without any measure of uncertainty. The evidence is therefore insufficient to support the headline improvement.
  3. [Section 4.2.2, Eq. (7)] The pertinence formula is undefined at a common input: if |N_P^t| = 1 (an element node with exactly one descendant text node having non-zero query score), the exponent |N_P^t|/(|N_P^t|-1) divides by zero. This is not an exotic configuration. For |N_P^t| = 2 the multiplier is e^2 ≈ 7.39, while for |N_P^t| = 3 it drops to e^1.5 ≈ 4.48; the formula therefore has a discontinuity that strongly privileges nodes with very few matching descendants. No justification for this behavior is given, and the ranking can become ill-defined precisely for simple leaf-parent nodes.
  4. [Section 5.1, Table 2] Precision and recall are not defined at the fragment level. It is not stated whether the returned unit is a text node, element node, or document, how overlapping element results are counted, or what the relevance-judgment unit is. The counts in Table 2 ('30 less specific docs returned including 15 relevant') presuppose a binary relevant/not-relevant decision for each returned fragment, but the protocol for making that decision is absent. Without this definition, the reported metrics cannot be compared across systems or reproduced.
minor comments (7)
  1. [Section 1] There is a typo in the discussion of markup formats: 'Y AML' should be 'YAML'.
  2. [Section 3.1, Figure 4a] The multiple-parent coefficient rule is unclear. Field's coefficient is shown as 2.5 = (3+2)/2, but the parent relationships in Figure 2 do not make the 3 and 2 terms obvious. Clarify the propagation order and which parents are averaged.
  3. [Section 4.1.1, Eq. (5)] The profile update rule contains no normalization, so the center-of-interest weights can grow without bound across requests. If only relative ordering matters, state this explicitly and also specify how the query weight w_tj is computed.
  4. [Section 4.2.1] The query concept vector construction is described only verbally ('according to the type of concepts sought'). Provide a concrete weighting formula for w_tj so that the experiments are reproducible.
  5. [Section 5.1] The introduction promises that 'source code and experimental data' will be made publicly available, but no repository link or data-availability statement is provided in the manuscript.
  6. [Figure 14] The caption text contains a long encoding artifact (a sequence of '/uni0000...' strings), making the caption unreadable.
  7. [References] Several references are not clearly connected to the claims they support (e.g., [3] on multivariate representation learning, [10] on distributed databases). Tightening the citation-to-claim mapping would help the reader.

Circularity Check

2 steps flagged · score 8.0 of 10

Central effectiveness claim is circular: the test corpus, queries, and relevance counts are all constructed from the same domain ontology that drives indexing, weighting, and profile reinforcement.

  1. self definitional [Section 5.1 (Settings); results Tables 2–3]
    "we built our own test database. It contains 400 XML documents covering several computer science topics, organized according to the concepts of our domain ontology. We also generated 300 queries by selecting ontology concepts and using their keywords to simulate different information needs."

    The retrieval machinery (Eq. 4) weights concepts by tf·iec f·W_R and (Eq. 6) propagates them in the XML tree; the test collection is 'organized according to the concepts of our domain ontology,' queries are 'generated by selecting ontology concepts,' and Table 2's relevant-document counts are given without any independent relevance protocol or external benchmark. Thus 'relevant' is, in this paper, synonymous with 'associated with the ontology concept used in the query'—the very relation the index, weighting, and cosine matching are built to rank. The reported precision/recall therefore measures the system's agreement with its own concept labeling, so the claimed gain over keyword retrieval is forced by the evaluation construction rather than demonstrated.

  2. self definitional [Section 4.1.1, Eq. (5); Section 5.1 and Table 3]
    "After each request, the weight of each center of interest is updated as: wCI_j(t+1)= (e^{w_tj} −1) + wCI_j(t) ... we report results (i) for repeated requests issued by the same user profile (to isolate the impact of ontology weighting and profile integration), and (ii) for multiple users with different interests over successive instants T1–T8."

    Eq. (5) adds (e^{w_tj}−1) to interest weight j whenever the query contains concept j, so the profile is literally fit to the query concepts; Eq. (6) multiplies every element-node weight by that interest vector. Table 3 then evaluates the same profile on repeated requests (T1–T8) over queries built from those exact ontology concepts, with relevance counts drawn from the same ontology-organized corpus. Improved scores after reinforcement are a mathematical consequence of the update rule plus the self-referential test bed: a concept queried once has its weight increased, making ontology-aligned documents more similar on the next repetition. The 'dynamic adaptation' result is therefore an artifact of the definitions, not independent evidence.

full rationale

Score is high because the central claimed result—that ontology weighting plus profile reinforcement improves precision/recall over keyword approaches—is supported only by an evaluation in which the corpus, query set, and relevance counts are generated from the same domain ontology that supplies the index vectors, weights, and profile terms. Section 5.1 explicitly constructs the 400-document collection 'organized according to the concepts of our domain ontology' and generates queries 'by selecting ontology concepts'; Tables 2–3 provide relevance counts without any external assessor or protocol. The improvement is therefore a measure of the system's self-consistency. Eq. (5) compounds this: the profile vector is updated by adding query-concept weights, and repeated-request experiments then report gains caused by the update. A separate weakness (not circularity) is that the baseline is 'as in Ounnaci et al., 2014'—the authors' own prior ontology method—while the abstract promises comparison with 'keyword-based approaches'; no keyword system is actually run. The mathematical formulas for weighting and similarity are not themselves circular, but they only measure similarity in the same ontology space, so they do not break the evaluation loop.

Assumptions & free parameters 3 free parameters · 4 assumptions · 0 invented entities

No new physical or mathematical entities are introduced. The free parameters are the hand-designed coefficient scheme, the exponential profile-update function, and the arbitrary exponent factors in the pertinence formula. The principal axioms are the ontology's adequacy, the unstated concept-extraction step, the validity of the self-built evaluation set, and the assumption that depth correlates with informativeness.

free parameters (3)
  • Ontology concept coefficients Coef(C_k) = Domain=1, Field=2.5, Path=2, Element=3, Script=4, Concept=5, Granule=4.5
    Hand-assigned depth coefficients (with averaging for multiple parents) determine the entire concept-weighting via Eq. (3); the choice encodes the prior that deeper nodes are more informative and is not fitted to external evidence.
  • Exponential profile update rule = wCI_j(t+1) = (e^{w_tj} - 1) + wCI_j(t)
    Ad hoc reinforcement of profile weights; has no normalization and grows unboundedly, and no learning rate or decay is specified; affects ranking through Eq. (6).
  • Pertinence formula exponent factors = e^{...} factors, base e
    The functional form of the relevance boost/penalty in Eq. (7) is chosen without derivation; the expression is ambiguous and undefined when the number of positive-scoring descendants equals 1.
assumptions (4)
  • domain assumption The domain ontology (Domain/Path/Field/Element/Script/Concept/Granule) is an adequate representation of the computer-science domain.
    Inherited from OBIREX [50] and [49]; the framework's vector space is defined over this ontology.
  • domain assumption Ontology concepts can be identified in XML text nodes by keyword occurrence counts c_f_ij.
    Eq. (4) requires counting concept occurrences in text nodes, but the concept-extraction/annotation algorithm is never described.
  • ad hoc to paper The self-built 400-document corpus, 300 queries, and relevance labels constitute a valid test of personalized retrieval effectiveness.
    Corpus organized by the ontology, queries generated from its concepts; no external benchmark or independent relevance assessment; this assumption is load-bearing for the experimental claim.
  • ad hoc to paper Deeper ontology concepts are more useful for retrieval than shallower ones.
    The paper states lower-level nodes 'provide more precise and targeted information' and builds the weighting on it, but offers no empirical or theoretical justification.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Ontology-driven personalized information retrieval for XML documents." pith.science (2026). https://pith.science/paper/77OZXZE5

@misc{pith2026260321139,
  author       = {Pith},
  title        = {Pith review of: Ontology-driven personalized information retrieval for XML documents},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/77OZXZE5}},
  note         = {Machine review of arXiv:2603.21139}
}
read the original abstract

This paper addresses the challenge of improving information retrieval from semi-structured eXtensible Markup Language (XML) documents. Traditional information retrieval systems (IRS) often overlook user-specific needs and return identical results for the same query, despite differences in users' knowledge, preferences, and objectives. We integrate external semantic resources, namely a domain ontology and user profiles, into the retrieval process. Documents, queries, and user profiles are represented as vectors of weighted concepts. The ontology applies a concept-weighting mechanism that emphasizes highly specific concepts, as lower-level nodes in the hierarchy provide more precise and targeted information. Relevance is assessed using semantic similarity measures that capture conceptual relationships beyond keyword matching, enabling personalized and fine-grained matching among user profiles, queries, and documents. Experimental results show that combining ontologies with user profiles improves retrieval effectiveness, achieving higher precision and recall than keyword-based approaches. Overall, the proposed framework enhances the relevance and adaptability of XML search results, supporting more user-centered retrieval.

Figures

Figures reproduced from arXiv: 2603.21139 by the authors.

Figure 1
Figure 1. A taxonomy of related work on XML information retrieval and personalization. [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Knowledge division of the domain ontology. [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Extract from the computer science domain ontology. [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (11 more)
Figure 4
Figure 4. Figure 4: Illustration of concept coefficient assignment and the resulting concept weights. Next, we compute the margin (∆), which is used to determine the final weights of ontology concepts. The margin is defined as: ∆ = 1 P|N| k=1 (Coef(Ck) − Coef(C1)) 2 (1) where: • 1 is th…
Figure 6
Figure 6. Figure 6: Representation of an XML document as a tree. Path 0.1378 Domain 0.1334 Field 0.1400 Element 0.1422 Script 0.1466 Concept 0.1510 Granule 0.1488 XML Document <Doc> <Title> European Museum Guide </Title> <Sect1> <Part1> The Louvre Museum is located in Paris city and Paris…
Figure 6
Figure 6. Figure 6: Concept propagation from text nodes to element nodes for index construction. [PITH_FULL_IMAGE:figures/full_fig_p010_6.png]
Figure 7
Figure 7. Figure 7: Example of start and end values assigned to nodes in an XML document. [PITH_FULL_IMAGE:figures/full_fig_p011_7.png]
Figure 9
Figure 9. Figure 9: Representation of the doc.xml document in a two-dimensional space based on the begin and end coordinates [PITH_FULL_IMAGE:figures/full_fig_p012_9.png]
Figure 10
Figure 10. Figure 10: Example of extraction of the descriptors of a node. <Title, 2, 3, 1> <Part2, 7, 8, 4> Part2 Title Part1 Sect1 Doc 1 10 4 9 2 3 5 6 7 8 <Doc 1 > <Title 2 > </Title 3 > <Sect1 4 > <Part1 5 > </Part1 6 > <Part2 7 > </Part2 8 > </Sect1 9 > </Doc 10> 2 1 1 4 7 Ancestors N.…
Figure 10
Figure 10. Figure 10: General architecture of the developed prototype. [PITH_FULL_IMAGE:figures/full_fig_p014_10.png]
Figure 11
Figure 11. Figure 11: Distribution of documents across domains in the XML collection. [PITH_FULL_IMAGE:figures/full_fig_p015_11.png]
Figure 12
Figure 12. Figure 12: Most frequent ontology concepts in the XML collection. [PITH_FULL_IMAGE:figures/full_fig_p016_12.png]
Figure 13
Figure 13. Figure 13: Word cloud of the XML collection content. [PITH_FULL_IMAGE:figures/full_fig_p016_13.png]
Figure 14
Figure 14. Figure 14: Clustermap of documents based on content similarity. [PITH_FULL_IMAGE:figures/full_fig_p017_14.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

64 extracted references · 4 linked inside Pith

  1. [1]

    Salton, Automatic information organization and retrieval., McGraw Hill Text, 1968

    G. Salton, Automatic information organization and retrieval., McGraw Hill Text, 1968

    1968

  2. [2]

    Abualigah, A

    L. Abualigah, A. H. Gandomi, M. A. Elaziz, A. G. Hussien, A. M. Khasawneh, M. Alshinwan, E. H. Houssein, Nature-inspired optimization algorithms for text document clustering—a comprehensive analy- sis, Algorithms 13 (12) (2020) 345

    2020

  3. [3]

    Zamani, M

    H. Zamani, M. Bendersky, Multivariate representation learning for information retrieval, in: Proceedings of the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval, 2023, pp. 163–173

    2023

  4. [4]

    H. Safi, M. Jaoua, L. H. Belguith, Pirat: a personalized information retrieval system in arabic texts based on a hybrid representation of a user profile, in: International Conference on Applications of Natural Language to Information Systems, Springer, 2016, pp. 326–334

    2016

  5. [5]

    Yahia, S

    N. Yahia, S. A. Mokhtar, A. Ahmed, Automatic generation of owl ontology from xml data source, arXiv preprint arXiv:1206.0570 (2012)

    arXiv 2012

  6. [6]

    Purificato, L

    E. Purificato, L. Boratto, E. W. De Luca, User modeling and user profiling: A comprehensive survey, arXiv preprint arXiv:2402.09660 (2024)

    arXiv 2024

  7. [7]

    J. H. Brenas, E. K. Shin, A. Shaban-Nejad, Adverse childhood experiences ontology for mental health surveillance, research, and evaluation: advanced knowledge representation and semantic web techniques, JMIR mental health 6 (5) (2019) e13498

    2019

  8. [8]

    Boneva, J

    I. Boneva, J. E. Labra Gayo, E. G. Prud’Hommeaux, Semantics and validation of shapes schemas for rdf, in: International semantic web conference, Springer, 2017, pp. 104–120

    2017

Show all 64 references
  1. [9]

    Neveditsin, P

    N. Neveditsin, P. Lingras, V . K. Mago, Evaluating structured output robustness of small language models for open attribute-value extraction from clinical notes, in: Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (V olume 4: Student Res...

    2025

  2. [10]

    M. T. Özsu, P. Valduriez, et al., Principles of distributed database systems, V ol. 2, Springer, 1999

    1999

  3. [11]

    K. A. Hambarde, H. Proenca, Information retrieval: recent advances and beyond, IEEE Access 11 (2023) 76581–76604

    2023

  4. [12]

    M. S. Kim, Y . H. Kong, Ontology-dtd matching algorithm for efficient xml query, in: International Con- ference on Fuzzy Systems and Knowledge Discovery, Springer, 2005, pp. 1093–1102

    2005

  5. [13]

    Chiaramella, Information retrieval and structured documents, in: European Summer School on Infor- mation Retrieval, Springer, 2000, pp

    Y . Chiaramella, Information retrieval and structured documents, in: European Summer School on Infor- mation Retrieval, Springer, 2000, pp. 286–309

    2000

  6. [14]

    Zargayouna, S

    H. Zargayouna, S. Salotti, Mesure de similarité dans une ontologie pour l’indexation sémantique de doc- uments xml, in: 15èmes Journées francophones d’Ingénierie des Connaissances, Presses universitaires de Grenoble, 2004, pp. 249–260

    2004

  7. [15]

    Harrathi, S

    R. Harrathi, S. Calabretto, Une approche de recherche sémantique dans les documents semi-structurés, Atelier Recherche d’Information Sémantique, Marseille, France (2010)

    2010

  8. [16]

    Theobald, R

    M. Theobald, R. Schenkel, G. Weikum, Exploiting structure, annotation, and ontological knowledge for automatic classification of xml data., in: WebDB, 2003, pp. 1–6

    2003

  9. [17]

    Paz-Trillo, R

    C. Paz-Trillo, R. Wassermann, P. P. Braga, An information retrieval application using ontologies, Journal of the Brazilian Computer Society 11 (2) (2005) 17–31

    2005

  10. [18]

    N. Y . Ibrahim, S. A. Mokhtar, H. M. Harb, Towards an ontology based integrated framework for semantic web, arXiv preprint arXiv:1305.7058 (2013)

    2013 arXiv

  11. [19]

    Martin, E

    V . Martin, E. Bruno, E. Murisasco, Ontology driven indexing: Application to personalized information retrieval, in: International Conference on Database and Expert Systems Applications, Springer, 2014, pp. 242–256

    2014

  12. [20]

    Omotosho, C

    L. Omotosho, C. Akanbi, M. Ituen, I. Ogundoyin, An ontology-based information retrieval mechanism for e-learning management system, Journal of Computer Science and Its Application 26 (1) (2019)

    2019

  13. [21]

    M. A. Ullah, S. A. Hossain, Ontology-based information retrieval system for university: methods and reasoning, in: Emerging Technologies in Data Mining and Information Security: Proceedings of IEMIS 2018, V olume 3, Springer, 2018, pp. 119–128

    2018

  14. [22]

    R. Eito-Brun, Ontology-based information retrieval: Development of a semantic-based tool for the media industry, in: Research Conference on Metadata and Semantics Research, Springer, 2018, pp. 188–195

    2018

  15. [23]

    Ga¸ si, T

    A. Ga¸ si, T. Ensari, M. Da˘gtekin, Ontology supported semantic based image retrieval, Semantic Web 15 (5) (2024) 2125–2137

    2024

  16. [24]

    Baziz, M

    M. Baziz, M. Boughanem, H. Prade, Une approche de représentation de l’information en ri basée sur les sous-arbres, in: Conférence francophone en Recherche d’Information et Applications (CORIA 2007), Université de Saint-Etienne, 2007, pp. 335–350

    2007

  17. [25]

    Boubekeur, W

    F. Boubekeur, W. Azzoug, Concept-based indexing in text information retrieval, arXiv preprint arXiv:1303.1703 (2013)

    2013 arXiv

  18. [26]

    Kraft, F

    R. Kraft, F. Maghoul, C. C. Chang, Y! q: contextual search at the point of inspiration, in: Proceedings of the 14th ACM international conference on Information and knowledge management, 2005, pp. 816–823

    2005

  19. [27]

    Gauch, J

    S. Gauch, J. Chaffee, A. Pretschner, Ontology-based personalized search and browsing, Web intelligence and agent systems 1 (3-4) (2003) 219–234

    2003

  20. [28]

    Salton, et al., The smart system—experiments in automatic document processing, Prentice-Hall, En- glewood Cliffs, NJ (1971)

    G. Salton, et al., The smart system—experiments in automatic document processing, Prentice-Hall, En- glewood Cliffs, NJ (1971). 20

    1971

  21. [29]

    J. P. Mc Gowan, A multiple model approach to personalised information access, Ph.D. thesis, Citeseer (2003)

    2003

  22. [30]

    González, B

    G. González, B. López, J. L. De La Rosa, The emotional factor: An innovative approach to user modelling for recommender systems, in: Workshop on Recommendation and Personalization in e-Commerce, 2002, pp. 90–99

    2002

  23. [31]

    M. N. Uddin, T. H. Duong, V . Sean, G.-S. Jo, Construction of semantic user profile for personalized web search, in: International Conference on Computational Collective Intelligence, Springer, 2012, pp. 99–108

    2012

  24. [32]

    Kostadinov, La personnalisation de l’information, définition de modèle de profil utilisateur

    D. Kostadinov, La personnalisation de l’information, définition de modèle de profil utilisateur. rapport de dea, Ph.D. thesis, Master’s thesis, Université de Versailles, France (2003)

    2003

  25. [33]

    F. Liu, C. Yu, W. Meng, Personalized web search for improving retrieval effectiveness, IEEE Transactions on knowledge and data engineering 16 (1) (2004) 28–40

    2004

  26. [34]

    L. Fan, Q. Li, B. Liu, X.-M. Wu, X. Zhang, F. Lv, G. Lin, S. Li, T. Jin, K. Yang, Modeling user behavior with graph convolution for personalized product search, in: Proceedings of the ACM Web Conference 2022, 2022, pp. 203–212

    2022

  27. [35]

    R. Ying, R. He, K. Chen, P. Eksombatchai, W. L. Hamilton, J. Leskovec, Graph convolutional neural networks for web-scale recommender systems, in: Proceedings of the 24th ACM SIGKDD international conference on knowledge discovery & data mining, 2018, pp. 974–983

    2018

  28. [36]

    B. M. Fonseca, P. Golgher, B. Pôssas, B. Ribeiro-Neto, N. Ziviani, Concept-based interactive query ex- pansion, in: Proceedings of the 14th ACM international conference on Information and knowledge man- agement, 2005, pp. 696–703

    2005

  29. [37]

    Z. Gong, C. W. Cheang, L. Hou U, Multi-term web query expansion using wordnet, in: International conference on database and expert systems applications, Springer, 2006, pp. 379–388

    2006

  30. [38]

    Song, I.-Y

    M. Song, I.-Y . Song, X. Hu, R. B. Allen, Integration of association rules and ontologies for semantic query expansion, Data & Knowledge Engineering 63 (1) (2007) 63–75

    2007

  31. [39]

    Carpineto, G

    C. Carpineto, G. Romano, A survey of automatic query expansion in information retrieval, Acm Comput- ing Surveys (CSUR) 44 (1) (2012) 1–50

    2012

  32. [40]

    D. Zhou, X. Wu, W. Zhao, S. Lawless, J. Liu, Query expansion with enriched user profiles for personalized search utilizing folksonomy data, IEEE Transactions on Knowledge and Data Engineering 29 (7) (2017) 1536–1548

    2017

  33. [41]

    Audeh, P

    B. Audeh, P. Beaune, M. Beigbeder, La reformulation hybride des requêtes exploratoires à l’aide de con- cepts explicites et implicites, in: COnf’erence en Recherche d’Infomations et Applications-CORIA 2014, 11th French Information Retrieval Conference, 2014, pp. pp–247

    2014

  34. [42]

    S. E. Robertson, S. Walker, S. Jones, M. M. Hancock-Beaulieu, M. Gatford, et al., Okapi at TREC-3, British Library Research and Development Department, 1995

    1995

  35. [43]

    O. Choi, K. Kim, D. Wang, H. Yeh, M. Hong, Personalized mobile information retrieval system, Interna- tional Journal of Advanced Robotic Systems 9 (1) (2012) 11

    2012

  36. [44]

    Mimouni, M

    N. Mimouni, M. Fernàndez, A. Nazarenko, D. Bourcier, S. Salotti, A relational approach for information retrieval on xml legal sources, in: Proceedings of the Fourteenth International Conference on Artificial Intelligence and Law, 2013, pp. 212–216

    2013

  37. [45]

    Amer-Yahia, I

    S. Amer-Yahia, I. Fundulaki, L. V . Lakshmanan, Personalizing xml search in pimento, in: 2007 IEEE 23rd International Conference on Data Engineering, IEEE, 2006, pp. 906–915. 21

    2007

  38. [46]

    B. K. Meenakshi, C. Nalini, Survey on managing xml search through personalization, International Journal on Recent and Innovation Trends in Computing and Communication 3 (3) (2015) 1320–1322

    2015

  39. [47]

    Belkin, G

    N. Belkin, G. Muresan, X. Zhang, Using user’s context for ir personalization, in: Proceedings of the ACM/SIGIR Workshop on Information Retrieval in Context, 2004, pp. 23–25

    2004

  40. [48]

    On-At, Temporalité et réseaux sociaux: prise en compte de l’évolution dans la construction du profil utilisateur, Ph.D

    S. On-At, Temporalité et réseaux sociaux: prise en compte de l’évolution dans la construction du profil utilisateur, Ph.D. thesis, Université Paul Sabatier-Toulouse III (2017)

    2017

  41. [49]

    Iddir, A.-O

    O. Iddir, A.-O. Rachid, Information retrieval in educational structured documents adapted to learners needs, in: 2014 4th International Symposium ISKO-Maghreb: Concepts and Tools for knowledge Man- agement (ISKO-Maghreb), IEEE, 2014, pp. 1–8

    2014

  42. [50]

    Ahmed-Ouamer, A

    R. Ahmed-Ouamer, A. Hammache, Ontology-based information retrieval for e-learning of computer sci- ence, in: 2010 International Conference on Machine and Web Intelligence, IEEE, 2010, pp. 250–257

    2010

  43. [51]

    L. Wood, V . Apparao, L. Cable, M. Champion, M. Davis, J. Kesselman, T. Pixley, J. Robie, P. Sharpe, C. Wilson, Document object model (dom) level 2 specification, World Wide Web Consortium. www. w3. org/TR/DOM-Level-2 (2000)

    2000

  44. [52]

    W. A. Woods, Conceptual indexing: A better way to organize knowledge, Sun Microsystems, Inc., 1997

    1997

  45. [53]

    M. W. Berry, Z. Drmac, E. R. Jessup, Matrices, vector spaces, and information retrieval, SIAM review 41 (2) (1999) 335–362

    1999

  46. [54]

    Sparck Jones, A statistical interpretation of term specificity and its application in retrieval, Journal of documentation 28 (1) (1972) 11–21

    K. Sparck Jones, A statistical interpretation of term specificity and its application in retrieval, Journal of documentation 28 (1) (1972) 11–21

    1972

  47. [55]

    Robertson, H

    S. Robertson, H. Zaragoza, M. Taylor, Simple bm25 extension to multiple weighted fields, in: Proceedings of the thirteenth ACM international conference on Information and knowledge management, 2004, pp. 42– 49

    2004

  48. [56]

    Bouhini, M

    C. Bouhini, M. Géry, C. Largeron, Modèle de recherche d’information sociale centré utilisateur, in: Ex- traction et gestion des connaissances (EGC’2013), Hermann, 2013, pp. 275–286

    2013

  49. [57]

    Sauvagnat, M

    K. Sauvagnat, M. Boughanem, Propositions pour la pondération des termes et l’évaluation de la pertinence des éléments en recherche d’information structurée., in: CORIA, 2006, pp. 29–40

    2006

  50. [58]

    Salton, E

    G. Salton, E. A. Fox, H. Wu, Extended boolean information retrieval, Communications of the ACM 26 (11) (1983) 1022–1036

    1983

  51. [59]

    P. F. Dietz, Maintaining order in a linked list, in: Proceedings of the fourteenth annual ACM symposium on Theory of computing, 1982, pp. 122–127

    1982

  52. [60]

    Harrathi, S

    R. Harrathi, S. Calabretto, Un modèle pour l’interrogation visuelle des documents structurés., in: CORIA, 2007, pp. 211–222

    2007

  53. [61]

    N. Fuhr, N. Gövert, G. Kazai, M. Lalmas, INitiative for the Evaluation of XML Retrieval (INEX), Springer, 2002

    2002

  54. [62]

    E. M. V oorhees, D. K. Harman, et al., TREC: Experiment and evaluation in information retrieval, V ol. 63, MIT press Cambridge, 2005

    2005

  55. [63]

    Peters, M

    C. Peters, M. Braschler, P. Clough, Multilingual information retrieval: From research to practice, Springer, 2012

    2012

  56. [64]

    Ferro, C

    N. Ferro, C. Peters, Information Retrieval Evaluation in a Changing World, Springer, 2019. 22

    2019

Pith tools

Reviewed August 2, 2026 · model on record in the stance chip above.