Enterprises have to maintain a large collection of business process models. These models are annotated with additional semantic information to enhance the searching, comprehension and understanding of the models. However, such enhancement of the models is an time-consuming, error-prone and labor-intensive task. Therefore, several attempts have been made to develop techniques for annotating these models with additional semantic information. To the best of our knowledge, no attempt has been made for the identification and analysis of these annotation techniques which has thwarted the advancement of these techniques. To that end, this study has employed a systematic approach to identify a comprehensive set of annotation techniques. Secondly, a taxonomy of these annotation techniques is developed to classify these techniques based on their underlying annotation mechanism. Finally, an analysis and comparison of the automated and semi-automated techniques is performed. The study concludes that there is a need for developing the next generation of techniques that can automatically annotate process models with the semantic information. Software engineering, Business process modeling, Semantic annotation, Annotation techniques, ClassiBusiness organizations are continuously evolving as able number of studies have been conducted on the CEUR ISE 2022: 1st International workshop on Intelligent CEUR some additional semantic information. A consider-
1. Introduction
new business use cases are regularly defined and
the existing ones are modified to fulfill the
customer needs. Consequently, the underlying business
processes are changing and new processes are also
added [
quence of steps that are performed in a certain order to achieve a business goals. The formal representation of business process is called business process model which are designed using process modeling languages.
A majority of the stakeholders cannot search, comprehend and use process models due to the intricacies of process modeling languages. To address permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).
The Boolean expression (”Process” OR ”Business Process” OR ”Process Model” OR ”Business Process Model”) AND (”Annotation” OR ”Tagging” OR ”Taxonomy” OR ”Repository” OR ”Classification” OR ”Categorization” OR ”Categorisation” OR ”Labelling” OR ”Grouping”) facility to export all the search results to a spreadsheet. The search results of the electronic databases were exported in a single go, whereas the search results of Taylor and Francis were exported one-byone.
For the relevance assessment of the searched studies, Inclusion (I) and Exclusion (E) criteria and a screening procedure are defined. The criteria are as follows. • RQ2. What are the diferent types of techniques used for the annotation of process models? • RQ3. What are the strengths and weaknesses of process model annotation techniques?
Section 2 discusses the systematic approach used for searching and screening of relevant literature along with extraction and coding of annotation techniques. Section 3 presents the taxonomy of the annotation techniques that is proposed in this study. An analysis of the techniques is presented in Section 4. Finally, Section 5 concluded this study.
This section presents the first contribution, identification of annotation techniques. It is a three-phase procedure that is inspired by the Kitchenham’s guidelines for conducting a systematic literature review [2]. In particular, it includes searching for literature, and employing relevance screening for identifying the relevant studies, and extracting and coding of the relevant annotation techniques for the development of an artifact. The details of of the procedure are as follows.
formed searching through multiple digital libraries, which includes IEEE Xplore, ACM Digital Library, SpringerLink, Sciencedirect, Taylor Francis Online and Emerald Insight. The primary search is limited to peer-reviewed journal articles, conference and workshop proceedings. That is, white papers, newspaper articles and posters, are excluded from the search. Furthermore, the search is limited to the English articles published since the year 2002.
For the searching, relevant search terms and their combinations are used. The search terms are, business, process, model, and annotation. Additionally, 2.3. Screening procedure query expansion is performed by adding synonyms Screening procedure is composed of three steps. In and other related words. Table 1 presents complete the first step, a single team member performed Boolean search string. For performing the search, screening based on title and keyword of each study advance-searching features of electronic databases and marked it as relevant, irrelevant or ambiguous. are used. In some cases, search strings are cus- During this screening step, a study was marked as tomised to fulfill the specific requirements of the relevant or irrelevant based on a sound reasons. For target database. As search string query provides the quality assurance, the second team evaluated results in large quantity therefore only the top 1000 the correctness of relevant and irrelevant studies relevant references are considered. All electronic based on a random selection. In the second step, databases, except Taylor and Francis, provide the • Does the study discuss the mechanism of annotating of process models or its elements? (I) • Does study annotate models or its elements
with predefined labels? (I) • Does it annotate models with run-time gen
erated categories? (I) • Does study annotate with domain ontologies
with predefined concepts? (I) • Does study annotate with semantic similarity
measurement? (I) • Does study discuss any mechanism regarding business process models except their annotation like business process model verification, similarity and generation etc.? (E) • Does study conducted survey or interview only regarding business process model annotation? (E) relevance screening was performed in the same manner based on the abstract, summary or conclusion of the studies. In third step, full text of the shortlisted studies was identified and the relevance assessment step was repeated based on the complete content.
Any disagreement between the decisions of both team members were resolved with the consensus.
As a result of the literature search and screening procedure 56 relevant studies were identified that are used in the rest of the study.
we extracted some specific details about the annotation techniques in order to understand their mechanism. These following nuggets of information were extracted. Automatic
Semi-Automatic • The mechanism used for annotation of pro- Question Based cess models, clustering based, rule-based, etc. • The level at which the annotation is performed, process model level, elements level or both. Figure 1: Taxonomy of the annotation techniques. • The mechanism used for defining the anno
tation concepts, predefined or runtime. • The annotation concepts used by the tech- instance, [3] proposed a technique that takes innique, generic annotation concepts, domain- put labels of a process model element as input and specific or both. assigns it a labelling style without any human in• The level of automation of the annotation tervention. In contrast, the techniques that provide technique, automatic, semi-automatic or a list of recommendations for annotating process manual. models, and requires human experts to select the • Implementation of the proposed artifacts, if most appropriate annotation value are categorized available. as semi-automatic annotation techniques. The cases in which humans are required to perform the anno
Two researchers independently extracted the in- tation manually are classified as Manual annotation formation discussed above. The results of the data techniques. extraction were recorded and conflicts were resolved For the second-level classification, the automatic by the consensus approach. In some cases, the annotation techniques are categorized into three results were retraced back to full-text to develop sub-categories, Natural Language Processing (NLP) consensus. Accordingly, the generated information based, Rule based and Clustering based. The techwas generated for use in the rest of the study. niques that automatically annotate process models using the text processing tools are categorized as NLP based annotation techniques. The techniques 3. Taxonomy of the Annotation that automatically annotate the diferent parts of Techniques process models based on predefined rules are categorized as rule based annotation techniques. The Figure 1 presents the taxonomy of annotation tech- annotation techniques that use clustering to group niques for process model annotation that we have process models for the annotation are referred to as developed. It can be observed from the figure that clustering techniques. annotation techniques are firstly classified based on The semi-automatic annotation techniques are the level of automation, Automatic, Semi-automatic further classified into three sub-categories: similarand Manual. A techniques that annotates process ity based, ontology based and social tagging based models without any human involvement are clas- annotation techniques. The techniques that autosified as Automatic annotation techniques. For matically generate recommendations using textual for development of Semi-automatic and Automatic annotation techniques. From the distribution of Semi-Automatic category it can be observed that most of the techniques use Similarity based annotation techniques, whereas for the Automatic techniques, researchers mainly focused on NLP based annotation techniques which merely annotates process model elements.
the annotation techniques. In particular, we focus on two types of techniques, automated techniques and semi-automated techniques. The analysis and Grand Total 168 comparison is based on the five criteria, Annotation
Mechanism, Level of Annotation, Annotation Concepts, Type of technique and Implementation of the similarity techniques are categorized as similarity proposed solution. Where, Mechanism represents based techniques. The annotation techniques that that the proposed technique belongs to in which automatically generate ranking list of most related sub-category of taxonomy. Level of Annotation ontology concept are categorized as ontology based represents that the proposed technique performs semi-automatic annotation techniques. Whereas, annotations to a process model or elements of the the annotation techniques that provide a social plat- process model. Concepts list represents that the form to the users for assigning tags to process model proposed technique uses a predefined list of annotausing their knowledge, and afterwards uses machine tion concepts or it generates new labels at run-time. learning techniques to automatically annotate pro- The fourth criterion, type of technique, represents cess models are categorized as social tagging based whether the proposed approach uses a supervised semi-automatic annotation techniques. approach or not, whereas the fifth criterion rep
The manual annotation techniques are further resents whether a prototype/tool is developed to classified into four sub-categories: Domain Expert demonstrate the annotation of process models. based, Modeling Expert based, Community based annotation and Question Based annotation. The 4.1. Automatic Techniques annotation techniques in which domain experts manually annotate process models are categorized as Table 3 presents the analysis and comparison of audomain expert based techniques. While, the tech- tomatic process model annotation techniques. The niques in which modeling experts manually annotate ifrst notable observation is that a majority of techprocess models are grouped into modeling expert niques employs an NLP based mechanism for the based annotation techniques. The annotation tech- annotation, whereas little attention has been paid to niques in which a community or a group of people the development of Clustering and Rule based techannotate process models in a controlled environ- niques. One possible reason stems from the maturity ment are classified as community based annotation of NLP discipline. The second notable observation techniques. Whereas, the annotation techniques in is that a majority of the annotation techniques which an ordinary person can annotate process mod- perform annotations at repository level, meaning els by answering the predefined questions are cate- that an annotation is assigned to a complete progorized as question based annotation techniques. cess model. Furthermore, it can be observed that
Table 2 presents the distribution of the anno- all the Clustering and Rule based techniques protation techniques. It can be observed from the pose repository level annotations. In contrast, NLP table that a vast majority of the annotation tech- based techniques perform annotation at repository niques falls in the Manual category, whereas there is level as well as at the model level. It is pertinent scarcity of Semi-automatic and Automatic annota- to mention that no efort has been made to protion techniques. This indicates that there is a need pose such technique that performs both model and repository level annotations.
The third notable observation is that a large ma- annotation of process models. The second notable jority of automatic annotation techniques use prede- observation is that a small majority of the annotaifned classes for the annotation, whereas some tech- tion techniques perform annotations at model level, niques generates annotation information at the run- i.e. annotations are performed with the element of time. A further analysis of the techniques revealed process model. Furthermore, majority of Ontology that all the NLP based techniques use predefined based techniques perform annotations at the model concepts for the annotation, whereas the Clustering level. Whereas, Similarity based techniques pertechniques used runtime generated concepts for the forms annotation at repository level, as well as at annotation. It can also be observed from the ta- the model level. It is also pertinent to mention that ble that none of the annotation techniques employ no efort has been made to propose a semi-automatic supervised learning techniques for the annotation, technique that performs element and model level although several research domains have benefited annotations. from the supervised learning techniques. Therefore, The third notable observation is that a large mawe recommended to propose supervised learning jority of semi-automatic annotation techniques use techniques for the annotation. Finally, it can be predefined classes for the annotation, whereas some observed that merely seven automatic annotation techniques generates annotation information at the techniques are evaluated practically by developing runtime. In contrast to automatic annotation techa prototype applications, whereas, all the remain- niques, there are three semi-automatic techniques ing eleven automatic annotation techniques merely that deals with both predefined and run-time gendemonstrated the efectiveness using illustration. erated annotation concepts. A further analysis of the techniques revealed that all Ontology based 4.2. Semi-automatic Techniques techniques have used predefined concepts for the annotation, whereas the majority of Social Tagging Table 4 presents the analysis and comparison of based techniques have used runtime generated. semi-automatic process model annotation tech- It can also be observed from the table that none of niques. The first notable observation is that a the semi-automatic annotation technique employs majority of techniques employs Similarity based supervised learning approach for the annotation. mechanism for the annotation, whereas little focus Moreover, it can be observed that in contrast to has been placed to the development of Ontology and automatic annotation techniques, a large majority Social Tagging based techniques. It shows the ma- of semi-automatic annotation techniques are evaluturity of Similarity domain for the semi-automatic ated practically by developing a prototype, whereas, eight semi-automatic annotation techniques merely mated and semi-automated techniques is presented. illustrated the use of the proposed techniques. It The analysis of the results revealed the existence shows the maturity of semi-automatic annotation of research gap for the partial or fully automation techniques regarding their practical evaluation. Fi- of annotation techniques. One notable gap is that nally, further analysis of Similarity based annota- little attention has been paid on the development of tion techniques revealed that recommendation based an automated or even semi-automated techniques. semi-automatic annotation techniques can be auto- Secondly, there are vast opportunities to benefit mated by employing a ranking or other statistical from the advancements of machine learning techprocedures. niques to develop fully automated techniques. Also, there is a need for developing next generation of annotation techniques that can perform annotation 5. Conclusion at model level, as well as at the element level. In this study, we have employed a systematic procedure to identify 56 studies that proposed techniques for the annotation of process models. Subsequently, a taxonomy of the annotation techniques has been developed by employing a bottom up approach. The taxonomy can serve as a reference for research community of this domain. Lastly, an analysis of autotative ordering relations, Information Sciences tems Modeling, volume 248 of Lecture Notes in 495 (2019) 100–115. doi:10.1016/j.ins.2019. Business Information Processing, Springer In04.058. ternational Publishing, Cham, 2016, pp. 51–65. [2] B. Kitchenham, Procedures for performing sys- doi:10.1007/978-3-319-39429-9_4. tematic reviews, Keele University Technical [9] F. Gao, S. Bhiri, Capability annotation of Report TR/SE-0401, Keele University, Keele, actions based on their textual descriptions, UK, 2004. in: Proceedings of the IEEE 23rd. Interna[3] H. Leopold, Parsing and annotating pro- tional WETICE Conference, IEEE, Parma, cess model elements, in: H. Leopold (Ed.), Italy, 2014, pp. 257–262. doi:10.1109/WETICE. Natural Language in Business Process Mod- 2014.68. els, volume 168 of Lecture Notes in Busi- [10] H. Leopold, C. Meilicke, M. Fellmann, F. Pitness Information Processing, Springer Inter- tke, H. Stuckenschmidt, J. Mendling, Towards national Publishing, Cham, 2013, pp. 49–80. the automated annotation of process models, doi:10.1007/978-3-319-04175-9_3. in: J. Zdravkovic, M. Kirikova, P. Johannes[4] D. Ferreira, M. Zacarias, M. Malheiros, P. Fer- son (Eds.), Proceedings of the 27th. Internareira, Approaching process mining with se- tional Conference on Advanced Information quence clustering: Experiments and findings, Systems Engineering, volume 9097 of Lecture in: G. Alonso, P. Dadam, M. Rosemann Notes in Computer Science, Springer Interna(Eds.), Proceedings of the 5th. International tional Publishing, Cham, 2015, pp. 401–416. Conference on Business Process Management, doi:10.1007/978-3-319-19069-3_25. BPM’07, volume 4714 of Lecture Notes in [11] A. Bögl, M. Schrefl, G. Pomberger, N. WeComputer Science, Springer Berlin Heidel- ber, Semantic annotation of epc models berg, Berlin, Heidelberg, 2007, pp. 360–374. in engineering domains to facilitate an audoi:10.1007/978-3-540-75183-0_26. tomated identification of common modelling [5] D. R. Ferreira, Applied sequence clustering practices, in: J. Filipe, J. Cordeiro (Eds.), techniques for process mining, in: J. Cardoso, Proceedings of the 10th. International ConW. v. d. Aalst (Eds.), Handbook of Research on ference on Enterprise Information Systems, Business Process Modeling, IGI Global, 2009, volume 19 of Lecture Notes in Business Inpp. 481–502. doi:10.4018/978-1-60566-288-6. formation Processing, Springer Berlin Heidelch022. berg, Berlin, Heidelberg, 2009, pp. 155–171. [6] M. Malinova, R. Dijkman, J. Mendling, Au- doi:10.1007/978-3-642-00670-8_12. tomatic extraction of process categories from [12] C. Figueroa, H. Ordoñez, J.-C. Corrales, C. Coprocess model collections, in: N. Lohmann, bos, L. K. Wives, E. Herrera-Viedma, ImM. Song, P. Wohed (Eds.), Proceedings of the proving business process retrieval using cateInternational Conference on Business Process gorization and multimodal search, KnowledgeManagement, BPM’13 International Work- Based Systems 110 (2016) 49–59. doi:10.1016/ shops, volume 171 of Lecture Notes in Busi- j.knosys.2016.07.014. ness Information Processing, Springer Interna- [13] H. K. Dam, A. Ghose, Mining version histories tional Publishing, Cham, 2014, pp. 430–441. for change impact analysis in business process doi:10.1007/978-3-319-06257-0_34. model repositories, Computers in Industry 67 [7] A. Ordoñez, H. Ordoñez, J. C. Corrales, C. Co- (2015) 72–85. doi:10.1016/j.compind.2014.10. bos, L. K. Wives, L. H. Thom, Grouping of 005. business processes models based on an incre- [14] K. Hinge, A. Ghose, G. Koliadis, Process seer: mental clustering algorithm using fuzzy sim- A tool for semantic efect annotation of busiilarity and multimodal search, Expert Sys- ness process models, in: Proceedings of the tems with Applications 67 (2017) 163–177. 13th. International Conference on Enterprise doi:10.1016/j.eswa.2016.08.061. Distributed Object Computing, IEEE, 2009, [8] H. Leopold, H. van der Aa, F. Pittke, M. Raf- pp. 54–63. doi:10.1109/EDOC.2009.24. fel, J. Mendling, H. A. Reijers, Integrat- [15] M. Santiputri, A. K. Ghose, H. K. Dam, ing textual and model-based process descrip- Mining task post-conditions: Automating the tions for comprehensive process search, in: acquisition of process semantics, Data & R. Schmidt, W. Guédria, I. Bider, S. Guer- Knowledge Engineering 109 (2017) 112–125. reiro (Eds.), Proceedings of the International doi:10.1016/j.datak.2017.03.007. Conference on BPMDS and EMMSAD, Enter- [16] M. Missikof, M. Proietti, F. Smith, Querying prise, Business-Process and Information Sys- semantically enriched business processes, in: A. Hameurlain, S. W. Liddle, K.-D. Schewe, Semantics IX, volume 4601 of Lecture Notes X. Zhou (Eds.), Proceedings of the 22nd. Inter- in Computer Science, Springer Berlin Heinational Conference on Database and Expert delberg, Berlin, Heidelberg, 2007, pp. 1–38. Systems Applications, Part II, volume 6861 of doi:10.1007/978-3-540-74987-5_1. Lecture Notes in Computer Science, Springer [23] N. Guarino, Semantic matching: Formal onBerlin Heidelberg, Berlin, Heidelberg, 2011, pp. tological distinctions for information organi294–302. doi:10.1007/978-3-642-23091-2_25. zation, extraction, and integration, in: M. T. [17] Y. Lin, H. Ding, Ontology-based semantic an- Pazienza (Ed.), Proceedings of the SCIE’97, Innotation for semantic interoperability of pro- formation Extraction A Multidisciplinary Apcess models, in: Proceedings of the Inter- proach to an Emerging Information Technology, national Conference on Computational Intel- volume 1299 of Lecture Notes in Computer ligence for Modelling, Control and Automa- Science, Springer Berlin Heidelberg, Berlin, tion and International Conference on Intel- Heidelberg, 1997, pp. 139–170. doi:10.1007/ ligent Agents, Web Technologies and Inter- 3-540-63438-X_8. net Commerce, CIMCA-IAWTIC’06, volume 1, [24] X. Wang, N. Li, H. Cai, B. Xu, An ontoIEEE, 2005, pp. 162–167. doi:10.1109/CIMCA. logical approach for semantic annotation of 2005.1631259. supply chain process models, in: R. Meers[18] Y. Lin, D. Strasunskas, Ontology-based seman- man, T. Dillon, P. Herrero (Eds.), Proceedtic annotation of process templates for reuse., igs of the Confederated International Conin: T. Halpin, K. Siau, J. Krogstie (Eds.), ferences: CoopIS, IS, DOA and ODBASE, Proceedings of the 10th. Workshop on Evalu- On the Move to Meaningful Internet Systems, ating Modeling Methods for Systems Analysis OTM’10, Part I, volume 6426 of Lecture Notes and Design, EMMSAD’05, held in conjunctiun in Computer Science, Springer Berlin Heidelwith the 17th. Conference on Advanced Infor- berg, Berlin, Heidelberg, 2010, pp. 540–554. mation Systems, CAiSE’05, Porto, Portugal, doi:10.1007/978-3-642-16934-2_40. 2005, pp. 207–218. URL: https://ceur-ws.org/ [25] Y. Lin, A. Sølvberg, Goal annotation of proVol-363/paper19.pdf. cess models for semantic enrichment of pro[19] Y. Lin, Semantic annotation for process mod- cess knowledge, in: J. Krogstie, A. Opdahl, els: Facilitating process knowledge manage- G. Sindre (Eds.), Proceedings of the 19th. ment via semantic interoperability, Ph.D. the- International Conference on Advanced Inforsis, Norwegian University of Science and Tech- mation Systems Engineering, volume 4495 of nology, Trondheim, Norway, 2008. URL: http: Lecture Notes in Computer Science, Springer //hdl.handle.net/11250/249894. Berlin Heidelberg, Berlin, Heidelberg, 2007, pp. [20] F. Smith, D. Bianchini, Semi-automatic pro- 355–369. doi:10.1007/978-3-540-72988-4_25. cess composition via semantics-enabled sub- [26] I. Ciuciu, G. Zhao, J. Mülle, S. von Stackelprocess selection and ranking, in: R. Poler, berg, C. Vasquez, T. Haberecht, R. MeersG. Doumeingts, B. Katzy, R. Chalmeta man, K. Böhm, Semantic support for (Eds.), Proceedings of the I-ESA Conferences, security-annotated business process models, in: Enterprise Interoperability V, volume 5 of T. Halpin, S. Nurcan, J. Krogstie, P. Sofer, IESACONF, Springer London, London, 2012, E. Proper, R. Schmidt, I. Bider (Eds.), Propp. 177–187. doi:10.1007/978-1-4471-2819-9_ ceedings of the 12th. International Workshop 16. on Business Process Modeling, Development [21] F. Smith, D. Bianchini, Selection, rank- and Support, Proceedings of the 16th. Intering and composition of semantically enriched national Conference on Exploring Modeling business processes, Computers in Industry Methods for Systems Analysis and Design, En65 (2014) 1253–1263. doi:10.1016/j.compind. terprise, Business-Process and Information Sys2014.07.009. tems Modeling, volume 81 of Lecture Notes [22] F. Giunchiglia, M. Yatskevich, P. Shvaiko, in Business Information Processing, Springer Semantic matching: Algorithms and imple- Berlin Heidelberg, Berlin, Heidelberg, 2011, pp. mentation, in: S. Spaccapietra, P. Atzeni, 284–298. doi:10.1007/978-3-642-21759-3_21. F. Fages, M.-S. Hacid, M. Kifer, J. Mylopou- [27] B. Vazquez, A. Martinez, A. Perini, H. Estrada, los, B. Pernici, P. Shvaiko, J. Trujillo, I. Za- M. Morandini, Enriching organizational modihrayeu (Eds.), Proceedings of the 13th. In- els through semantic annotation, Procedia ternational Conference on Cooperative Infor- Technology 7 (2013) 297–304. doi:10.1016/j. mation Systems, CoopIS’05, Journal on Data protcy.2013.04.037. [28] C. Di Francescomarino, P. Tonella, Supporting for unstructured information, in: Proceedings ontology-based semantic annotation of busi- of the Ninth Mexican International Conferness processes with automated suggestions, in: ence on Artificial Intelligence, IEEE, 2010, pp. T. Halpin, J. Krogstie, S. Nurcan, E. Proper, 81–85. doi:10.1109/MICAI.2010.22. R. Schmidt, P. Sofer, R. Ukor (Eds.), Pro- [34] C. Diamantini, D. Potena, E. Storti, Clustering ceedings of the 10th. International Workshop of process schemas by graph mining techniques, on Business Process Modeling, Development in: G. Mecca, S. Greco (Eds.), Proceedings and Support, Proceedings of the 14th. Inter- of the Nineteenth Italian Symposium on Adnational Conference on Exploring Modeling vanced Database Systems, SEBD’11, Maratea, Methods for Systems Analysis and Design, En- Italy, 2011, pp. 49–56. terprise, Business-Process and Information Sys- [35] J. Melcher, D. Seese, Visualization and clustems Modeling, volume 29 of Lecture Notes tering of business process collections based on in Business Information Processing, Springer process metric values, in: Proceedings of the Berlin Heidelberg, Berlin, Heidelberg, 2009, pp. 10th. International Symposium on Symbolic 211–223. doi:10.1007/978-3-642-01862-6_18. and Numeric Algorithms for Scientific Com[29] C. Di Francescomarino, P. Tonella, Sup- puting, IEEE, Timisoara, Romania, 2008, pp. porting ontology-based semantic annotation 572–575. doi:10.1109/SYNASC.2008.37. of business processes with automated sugges- [36] H. J. Wang, H. Wu, Supporting process tions, International Journal of Information design for e-business via an integrated proSystem Modeling and Design 1 (2010) 59–84. cess repository, Information Technology and doi:10.4018/jismd.2010040104. Management 12 (2011) 97–109. doi:10.1007/ [30] M. Born, F. Dörr, I. Weber, User-friendly s10799-010-0076-z.
semantic annotation in business process mod- [37] H. Wu, M. Zubair, K. Maly, Harvesting social eling, in: M. Weske, M.-S. Hacid, C. Go- knowledge from folksonomies, in: Proceedings dart (Eds.), Proceedings of the International of the Seventeenth Conference on Hypertext Conference on Web Information Systems En- and Hypermedia, HYPERTEXT’06, Associagineering, volume 4832 of Lecture Notes in tion for Computing Machinery, New York, NY, Computer Science, Springer Berlin Heidel- USA, 2006, pp. 111–114. doi:10.1145/1149941. berg, Berlin, Heidelberg, 2007, pp. 260–271. 1149962.
doi:10.1007/978-3-540-77010-7_25. [38] S. A. Golder, B. A. Huberman, Usage patterns [31] M. Born, J. Hofmann, T. Kaczmarek, of collaborative tagging systems, Journal of M. Kowalkiewicz, I. Markovic, J. Scicluna, Information Science 32 (2006) 198–208. doi:10. I. Weber, X. Zhou, Supporting execution-level 1177/0165551506062337. business process modeling with semantic tech- [39] C. Marlow, M. Naaman, D. Boyd, M. Davis, nologies, in: X. Zhou, H. Yokota, K. Deng, Ht06, tagging paper, taxonomy, flickr, acaQ. Liu (Eds.), Proceedings of the 14th. In- demic article, to read, in: Proceedings of ternational Conference on Database Systems the Seventeenth Conference on Hypertext and for Advanced Applications, volume 5463 of Hypermedia, HYPERTEXT’06, Association Lecture Notes in Computer Science, Springer for Computing Machinery, New York, NY, Berlin Heidelberg, Berlin, Heidelberg, 2009, pp. USA, 2006, pp. 31–40. doi:10.1145/1149941. 759–763. doi:10.1007/978-3-642-00887-0_67. 1149949. [32] M. Born, J. Hofmann, T. Kaczmarek, [40] R.-H. Eid-Sabbagh, M. Kunze, M. Weske, An M. Kowalkiewicz, I. Markovic, J. Scicluna, open process model library, in: F. Daniel, I. Weber, X. Zhou, Semantic annotation and K. Barkaoui, S. Dustdar (Eds.), Proceedings composition of business processes with mae- of the International Conference on Business stro, in: S. Bechhofer, M. Hauswirth, J. Hof- Process Management, Business Process Manmann, M. Koubarakis (Eds.), Proceedings of agement Workshops, BPM’11, Part II, volume the 5th. European Semantic Web Conference, 100 of Lecture Notes in Business Information volume 5021 of Lecture Notes in Computer Processing, Springer, Berlin, Heidelberg, 2011, Science, Springer Berlin Heidelberg, Berlin, pp. 26–38. doi:10.1007/978-3-642-28115-0_4. Heidelberg, 2008, pp. 772–776. doi:10.1007/ 978-3-540-68234-9_56. [33] D. A. Rosso-Pelayo, R. A. Trejo-Ramírez,
M. Gonzalez-Mendoza, N. Hernandez-Gress, Business process mining and rules detection