Journal:Research on information retrieval model based on ontology - Revision history

Shawndouglas: Converted ombox to template

2021-10-06T18:44:05Z

Converted ombox to template

← Older revision		Revision as of 18:44, 6 October 2021
Line 19:		Line 19:
	\|download = [https://jwcn-eurasipjournals.springeropen.com/track/pdf/10.1186/s13638-019-1354-z https://jwcn-eurasipjournals.springeropen.com/track/pdf/10.1186/s13638-019-1354-z] (PDF)		\|download = [https://jwcn-eurasipjournals.springeropen.com/track/pdf/10.1186/s13638-019-1354-z https://jwcn-eurasipjournals.springeropen.com/track/pdf/10.1186/s13638-019-1354-z] (PDF)
	}}		}}
	{{~~ombox~~
	~~\| type = content~~		{{Ombox math}}
	~~\| style =~~
	\| text = This article contains rendered mathematical formulae. You ''may'' require the [https://chrome.google.com/webstore/detail/tex-all-the-things/cbimabofgmfdkicghcadidpemeenbffn?hl=en TeX All the Things] plugin for Chrome or the [https://addons.mozilla.org/en-US/firefox/addon/native-mathml/ Native MathML] add-on and [https://developer.mozilla.org/en-US/docs/Mozilla/MathML_Project/Fonts fonts] for Firefox if they don't render properly for you.
	}}

	==Abstract==		==Abstract==

Shawndouglas: /* Notes */ Fixed cat

2019-11-25T18:30:46Z

Notes: Fixed cat

← Older revision		Revision as of 18:30, 25 November 2019
Line 326:		Line 326:
	[[Category:LIMSwiki journal articles (with rendered math)]]		[[Category:LIMSwiki journal articles (with rendered math)]]
	[[Category:LIMSwiki journal articles on information retrieval]]		[[Category:LIMSwiki journal articles on information retrieval]]
	[[Category:LIMSwiki journal articles on sensor networks]]		[[Category:LIMSwiki journal articles on sensors and sensor networks]]

Shawndouglas: /* Abbreviations */

2019-02-13T01:19:32Z

Abbreviations

← Older revision		Revision as of 01:19, 13 February 2019
Line 293:		Line 293:
	==Abbreviations==		==Abbreviations==
	'''IR''': Information retrieval		'''IR''': Information retrieval

	'''VSM''': Vector space model		'''VSM''': Vector space model

Shawndouglas: Finished adding rest of content.

2019-02-13T01:19:14Z

Finished adding rest of content.

← Older revision		Revision as of 01:19, 13 February 2019
Line 210:		Line 210:

	The different threshold similarity value ζ is taken, in which ζ = 0.5 means sim(''S<sub>q</sub>'', ''S<sub>j</sub>'') ≥ 0.55. Every experiment counts retrieval documents set results \|''A''\|, ontology relevant documents \|''B''\|, and user query relevant document in the retrieval set \|''A'' ∩ ''B''\| to calculate the precision and recall rate. The result is shown in Table 1.		The different threshold similarity value ζ is taken, in which ζ = 0.5 means sim(''S<sub>q</sub>'', ''S<sub>j</sub>'') ≥ 0.55. Every experiment counts retrieval documents set results \|''A''\|, ontology relevant documents \|''B''\|, and user query relevant document in the retrieval set \|''A'' ∩ ''B''\| to calculate the precision and recall rate. The result is shown in Table 1.


	{\|		{\|
Line 283:		Line 282:
	\|}		\|}

			The precision rate improves with the threshold increasing. The precision rate reaches more than 99% when ζ = 0.6. However, the recall rate only reaches 74%, which means the query result lost the critical information.

			When ζ = 0.5, the recall rate maintains a higher rate while precision remains low. Because of the system search, all the documents have ontology which relates with a query. The ζ = 0.55 balance both the precision rate and recall rate.

			==Conclusion==
			In order to better satisfy users’ retrieval needs and optimize the performance of information retrieval, domain ontology is introduced into the information retrieval system. In this paper, an information retrieval model based on domain ontology is proposed. The system includes document processing and ontology document retrieval with the ontology server, information database, and query transition and retrieval agent modules. We present a genetic algorithm to calculate the optimum combination of weighted factors of word frequency. Based on the evaluation criterion, we applied the system to query documents and compare with expert lists. In the end, the genetic algorithm shortens the distance compared with simulated annealing, and the ontology retrieval model exhibits a better precision and recall rate to understand the users’ requirements.

			In the future, we wish to further implement an automatic or semi-automatic method such as data mining to an established ontology database to prevent the high difficulty in ontology establishment. And we may further implement modeling personalized query preferences and return retrieval results according to different user query demands.

			==Abbreviations==
			'''IR''': Information retrieval
			'''VSM''': Vector space model

			==Declarations==
			===Acknowledgements===
			This work is supported by the Science and Technology Research Project of the Department of Education of Jilin Province (Grant 201657).

			====Funding====
			The Science and Technology Research Project of Department of Education of Jilin Province (Grant 201657).

			====Availability of data and materials====
			The data are included in this published article.

			====Author’s contributions====
			The manuscript was written through contributions of the author. The author read and approved the final manuscript.

			====Author’s information====
			Binbin Yu: Ph.D. candidate, College of Computer Science and Technology, Jilin University. Lecturer, College of Information Technology and Media, Beihua University. His research interests include network security and so on.

			====Competing interests====
			The author declares that he has no competing interests.

	==References==		==References==

Shawndouglas: Saving and adding more.

2019-02-13T01:03:04Z

Saving and adding more.

Show changes

Shawndouglas: Saving and adding more.

2019-02-13T00:01:39Z

Saving and adding more.

← Older revision		Revision as of 00:01, 13 February 2019
Line 72:		Line 72:
	Here, ''p<sub>i</sub>'' = ''r<sub>i</sub>/r'', ''q<sub>i</sub>'' = (''f<sub>i</sub>'' − ''r<sub>i</sub>'')/(''f'' − ''r''), ''f'' refers to amount of documents in the training document set. ''r'' is the number of documents related to the user query in the training document set. ''f<sub>i</sub>'' represents a number of documents, including index term ''k<sub>i</sub>'' in the training document set. ''R<sub>i</sub>'' is the number of documents, including ''k<sub>i</sub>'' in ''r'' relation documents.		Here, ''p<sub>i</sub>'' = ''r<sub>i</sub>/r'', ''q<sub>i</sub>'' = (''f<sub>i</sub>'' − ''r<sub>i</sub>'')/(''f'' − ''r''), ''f'' refers to amount of documents in the training document set. ''r'' is the number of documents related to the user query in the training document set. ''f<sub>i</sub>'' represents a number of documents, including index term ''k<sub>i</sub>'' in the training document set. ''R<sub>i</sub>'' is the number of documents, including ''k<sub>i</sub>'' in ''r'' relation documents.

	The Okapi BM25 model is called BM25, which is an algorithm based on the probabilistic retrieval model. The Okapi BM25 model ~~[13~~, ~~14]~~ is a model developed from the probabilistic model that incorporates term frequency and length normalization. The local weights are computed as parameterized frequencies, including term frequency, document frequency, and global weights as RSJ weights. Local weights are based on a 2D Poisson model, while the global weights are based on the Robertson-Spärck-Jones Probabilistic Model. By reducing the number of parameters to be learned and approximated, based on these heuristic techniques, BM25 often achieves better performance compared to TF-IDF (term frequency–inverse document frequency).		The Okapi BM25 model is called BM25, which is an algorithm based on the probabilistic retrieval model. The Okapi BM25 model<ref name="VoorheesTREC05">{{cite book \|title=TREC: Experiment and Evaluation in Information Retrieval \|editor=Voorhees, E.M.; Harman, D.K. \|publisher=MIT Press \|pages=368 \|year=2005 \|isbn=9780262220736}}</ref><ref name="PereiraContext05">{{cite journal \|title=Contextual weighted representations and indexing models for the retrieval of HTML documents \|journal=Soft Computing \|author=Pereira, R.A.M.; Molinari, A.; Pasi, G. \|volume=9 \|issue=7 \|pages=481-92 \|year=2005 \|doi=10.1007/s00500-004-0361-z}}</ref> is a model developed from the probabilistic model that incorporates term frequency and length normalization. The local weights are computed as parameterized frequencies, including term frequency, document frequency, and global weights as RSJ weights. Local weights are based on a 2D Poisson model, while the global weights are based on the Robertson-Spärck-Jones Probabilistic Model. By reducing the number of parameters to be learned and approximated, based on these heuristic techniques, BM25 often achieves better performance compared to TF-IDF (term frequency–inverse document frequency).

	==Based on the domain ontology information retrieval model==		==Based on the domain ontology information retrieval model==
Line 91:		Line 91:
	\|}		\|}

			The system consists of two parts: ontology document processing (including domain ontology servers, data source, document process unit, and information database) and ontology document retrieval (including domain ontology server, query transition, custom process, and retrieval agent).

			===Ontology documents processing===
			Document processing extracts useful information from an unstructured text message and establishes mapping relations between document terms and concepts based on domain ontology.<ref name="ZhangResearch08">{{cite journal \|title=Research on ontology-based information retrieval system models \|journal=Application Research of Computers \|author=Zhang, K.; Nan, K.; Ma, Y. \|volume=8 \|issue=25 \|pages=2241-49 \|year=2008 \|url=https://www.oriprobe.com/journals/jsjyyyj/2008_8.html}}</ref> Document processing is shown in Fig. 2.


			[[File:Fig2 Yu JOnWireCommNet2019 2019.png\|600px]]
			{{clear}}
			{\|
			\| STYLE="vertical-align:top;"\|
			{\| border="0" cellpadding="5" cellspacing="0" width="600px"
			\|-
			\| style="background-color:white; padding-left:10px; padding-right:10px;"\| <blockquote>'''Fig. 2''' Ontology document processing</blockquote>
			\|-
			\|}
			\|}

			In the preprocessing procedure, each document in the document set implements vocabulary, analyzes words, and filters numbers, hyphens, and punctuation. Using a stop word list removes function words to leave useful words such as nouns and verbs.<ref name="KimAnEff15">{{cite journal \|title=An Efficient Sensor Deployment Scheme for Large-Scale Wireless Sensor Networks \|journal=IEEE Communications Letters \|author=Kim, H.; Han, S.-W. \|volume=19 \|issue=1 \|pages=98–101 \|year=2015 \|doi=10.1109/LCOMM.2014.2372015}}</ref> Extracting stem words and removing the prefix and suffix improve the accuracy of retrieval. Finally, determining certain words as an index element expresses literature content conception.

			Annotating semantic on a retrieved object by analyzing characteristic vocabulary builds the mapping relation between words and concepts. First, characteristic words are extracted and the weight of each word is calculated by counting word frequency to distinguish the importance of words. In this paper, the genetic algorithm is used to calculate the best weighting factor. In the end, it is applied to the actual retrieval system.

			The system automatically learns weighted factor by genetic algorithm. It is a heuristic method which simulates biological evolution processes and through factor mutation eliminates the non-ideal factor sets and leaves the optimal factor set. The algorithm tries to maximize the fitness function as a parameter estimation to search a population consisting of the fittest individual; in our case, those are the parameters of the weighted term in retrieving. In Fig. 3, the pseudo-code of genetic algorithm for weighted term frequency is described.


			[[File:Fig3 Yu JOnWireCommNet2019 2019.png\|350px]]
			{{clear}}
			{\|
			\| STYLE="vertical-align:top;"\|
			{\| border="0" cellpadding="5" cellspacing="0" width="350px"
			\|-
			\| style="background-color:white; padding-left:10px; padding-right:10px;"\| <blockquote>'''Fig .3''' Pseudo-code for select weight factor by genetic algorithm</blockquote>
			\|-
			\|}
			\|}

	==References==		==References==

Shawndouglas: Saving and adding more.

2019-02-12T23:33:04Z

Saving and adding more.

← Older revision		Revision as of 23:33, 12 February 2019
Line 57:		Line 57:

	<math>{\overset{\rightarrow}{d}}_{j} = \left( w_{1j},w_{2j},w_{3j},\ldots,w_{tj} \right)</math>		<math>{\overset{\rightarrow}{d}}_{j} = \left( w_{1j},w_{2j},w_{3j},\ldots,w_{tj} \right)</math>


	The Boolean model is a classical information retrieval (IR) model based on set theory and Boolean algebra. Boolean retrieval can be effective if a query requires unambiguous selection.<ref name="Baeza-YatesModern99">{{cite book \|title=Modern Information Retrieval \|author=Baeza-Yates, R.; Ribeiro-Neto, B. \|publisher=Addison Wesley \|year=1999 \|pages=544 \|isbn=9780201398298}}</ref> But it can only result in whether the document is related or not related. The Boolean model lacks the ability to describe the situation that query words partially match a paper. The similarity result of document ''d<sub>j</sub>'' and query ''q'' is binary, either 0 or 1. The binary value has limitations and the Boolean queries are hard to construct.		The Boolean model is a classical information retrieval (IR) model based on set theory and Boolean algebra. Boolean retrieval can be effective if a query requires unambiguous selection.<ref name="Baeza-YatesModern99">{{cite book \|title=Modern Information Retrieval \|author=Baeza-Yates, R.; Ribeiro-Neto, B. \|publisher=Addison Wesley \|year=1999 \|pages=544 \|isbn=9780201398298}}</ref> But it can only result in whether the document is related or not related. The Boolean model lacks the ability to describe the situation that query words partially match a paper. The similarity result of document ''d<sub>j</sub>'' and query ''q'' is binary, either 0 or 1. The binary value has limitations and the Boolean queries are hard to construct.
Line 67:		Line 68:

	<math>{Sim}\left( D,q \right) = \sum\log\frac{p_{i}\left( 1 - q_{i} \right)}{q_{i}\left( 1 - p_{i} \right)}</math>		<math>{Sim}\left( D,q \right) = \sum\log\frac{p_{i}\left( 1 - q_{i} \right)}{q_{i}\left( 1 - p_{i} \right)}</math>


	Here, ''p<sub>i</sub>'' = ''r<sub>i</sub>/r'', ''q<sub>i</sub>'' = (''f<sub>i</sub>'' − ''r<sub>i</sub>'')/(''f'' − ''r''), ''f'' refers to amount of documents in the training document set. ''r'' is the number of documents related to the user query in the training document set. ''f<sub>i</sub>'' represents a number of documents, including index term ''k<sub>i</sub>'' in the training document set. ''R<sub>i</sub>'' is the number of documents, including ''k<sub>i</sub>'' in ''r'' relation documents.		Here, ''p<sub>i</sub>'' = ''r<sub>i</sub>/r'', ''q<sub>i</sub>'' = (''f<sub>i</sub>'' − ''r<sub>i</sub>'')/(''f'' − ''r''), ''f'' refers to amount of documents in the training document set. ''r'' is the number of documents related to the user query in the training document set. ''f<sub>i</sub>'' represents a number of documents, including index term ''k<sub>i</sub>'' in the training document set. ''R<sub>i</sub>'' is the number of documents, including ''k<sub>i</sub>'' in ''r'' relation documents.

			The Okapi BM25 model is called BM25, which is an algorithm based on the probabilistic retrieval model. The Okapi BM25 model [13, 14] is a model developed from the probabilistic model that incorporates term frequency and length normalization. The local weights are computed as parameterized frequencies, including term frequency, document frequency, and global weights as RSJ weights. Local weights are based on a 2D Poisson model, while the global weights are based on the Robertson-Spärck-Jones Probabilistic Model. By reducing the number of parameters to be learned and approximated, based on these heuristic techniques, BM25 often achieves better performance compared to TF-IDF (term frequency–inverse document frequency).

			==Based on the domain ontology information retrieval model==
			The concept of domain ontology has a relation to other concepts simultaneously. The interrelation between concepts of the semantic relative network implements synonym expansion retrieval, semantic entailment expansion, and semantic correlation expansion. We introduce a domain ontology information retrieval model to apply ontology into the traditional information retrieval model by query expansion to improve efficiency.

			An illustration of the structure for the information retrieval model is shown in Fig. 1.


			[[File:Fig1 Yu JOnWireCommNet2019 2019.png\|600px]]
			{{clear}}
			{\|
			\| STYLE="vertical-align:top;"\|
			{\| border="0" cellpadding="5" cellspacing="0" width="600px"
			\|-
			\| style="background-color:white; padding-left:10px; padding-right:10px;"\| <blockquote>'''Fig. 1''' An illustration of structure for information retrieval mode.</blockquote>
			\|-
			\|}
			\|}

Shawndouglas at 23:02, 12 February 2019

2019-02-12T23:02:57Z

← Older revision		Revision as of 23:02, 12 February 2019
Line 57:		Line 57:

	<math>{\overset{\rightarrow}{d}}_{j} = \left( w_{1j},w_{2j},w_{3j},\ldots,w_{tj} \right)</math>		<math>{\overset{\rightarrow}{d}}_{j} = \left( w_{1j},w_{2j},w_{3j},\ldots,w_{tj} \right)</math>

			The Boolean model is a classical information retrieval (IR) model based on set theory and Boolean algebra. Boolean retrieval can be effective if a query requires unambiguous selection.<ref name="Baeza-YatesModern99">{{cite book \|title=Modern Information Retrieval \|author=Baeza-Yates, R.; Ribeiro-Neto, B. \|publisher=Addison Wesley \|year=1999 \|pages=544 \|isbn=9780201398298}}</ref> But it can only result in whether the document is related or not related. The Boolean model lacks the ability to describe the situation that query words partially match a paper. The similarity result of document ''d<sub>j</sub>'' and query ''q'' is binary, either 0 or 1. The binary value has limitations and the Boolean queries are hard to construct.

			The VSM, which is proposed earlier by Salton, is based on the vector space model theory and vector linear algebra operation, which abstract the query conditions and text into vectors in the multidimensional vector space. The multi-keyword matching here can express the meaning of the text more.<ref name="TangTheRes10" /> Compared with the Boolean model, the VSM calculates relevant document ranking by comparing the angle relating similarity between the vector of each document and the original query vector in the spatial representation.

			The probabilistic model<ref name="MaAConn16" /> mainly relies on probabilistic operation and Bayes rules to match data information. The probabilistic model not only considers the internal relations between keywords and documents, but it also retrieves texts based on probability dependency. The model, usually based on a group of parameterized probability distributions, consumes the internal relation between keywords and documents and retrieves according to probabilistic dependency. The model requires strong independent assumptions for tractability.

			The binary independence retrieval model<ref name="PremalathaText14">{{cite journal \|title=Text processing in information retrieval system using vector space model \|journal=Proceedings from the 2014 International Conference on Information Communication and Embedded Systems \|author=Premalatha, R.; Srinivasan, S. \|pages=1–6 \|year=2014 \|doi=10.1109/ICICES.2014.7033837}}</ref> is evolved from the probabilistic model with better performance. Assuming that document ''D'' and index term ''q'' is described in a two-valued vector (''x<sub>1</sub>'', ''x<sub>2</sub>'', … ''x<sub>n</sub>''), if index term ''k<sub>i</sub>'' ∈ ''D'', then ''x<sub>i</sub>'' = 1; otherwise, ''x<sub>i</sub>'' = 0. The correlation function of index term and document are shown below.

			<math>{Sim}\left( D,q \right) = \sum\log\frac{p_{i}\left( 1 - q_{i} \right)}{q_{i}\left( 1 - p_{i} \right)}</math>

			Here, ''p<sub>i</sub>'' = ''r<sub>i</sub>/r'', ''q<sub>i</sub>'' = (''f<sub>i</sub>'' − ''r<sub>i</sub>'')/(''f'' − ''r''), ''f'' refers to amount of documents in the training document set. ''r'' is the number of documents related to the user query in the training document set. ''f<sub>i</sub>'' represents a number of documents, including index term ''k<sub>i</sub>'' in the training document set. ''R<sub>i</sub>'' is the number of documents, including ''k<sub>i</sub>'' in ''r'' relation documents.



	==References==		==References==

Shawndouglas at 22:41, 12 February 2019

2019-02-12T22:41:22Z

← Older revision		Revision as of 22:41, 12 February 2019
Line 67:		Line 67:
	[[Category:LIMSwiki journal articles (added in 2019)‎]]		[[Category:LIMSwiki journal articles (added in 2019)‎]]
	[[Category:LIMSwiki journal articles (all)‎]]		[[Category:LIMSwiki journal articles (all)‎]]
			[[Category:LIMSwiki journal articles (with rendered math)]]
	[[Category:LIMSwiki journal articles on information retrieval]]		[[Category:LIMSwiki journal articles on information retrieval]]
	[[Category:LIMSwiki journal articles on sensor networks]]		[[Category:LIMSwiki journal articles on sensor networks]]

Shawndouglas at 22:20, 12 February 2019

2019-02-12T22:20:12Z

← Older revision		Revision as of 22:20, 12 February 2019
Line 22:		Line 22:
	\| type = content		\| type = content
	\| style =		\| style =
	\| text = This article contains rendered mathematical formulae. You ''may'' require the [https://chrome.google.com/webstore/detail/~~math~~-~~anywhere~~/~~gebhifiddmaaeecbaiemfpejghjdjmhc Math Anywhere~~] plugin for Chrome or the [https://addons.mozilla.org/en-US/firefox/addon/native-mathml/ Native MathML] add-on and [https://developer.mozilla.org/en-US/docs/Mozilla/MathML_Project/Fonts fonts] for Firefox if they don't render properly for you.		\| text = This article contains rendered mathematical formulae. You ''may'' require the [https://chrome.google.com/webstore/detail/tex-all-the-things/cbimabofgmfdkicghcadidpemeenbffn?hl=en TeX All the Things] plugin for Chrome or the [https://addons.mozilla.org/en-US/firefox/addon/native-mathml/ Native MathML] add-on and [https://developer.mozilla.org/en-US/docs/Mozilla/MathML_Project/Fonts fonts] for Firefox if they don't render properly for you.
	}}		}}