1
Daily Knowledge
From the beginning, knowledge has been a preoccupation for humans. A lot of questions are still being discussed: What is knowledge? How is knowledge built? How is it represented in the mind? How can it be kept? How can it be learned? etc.
In this book, we deal with the notion of daily knowledge. We try to answer questions that have been discussed before. However, first of all, let us present the notion of knowledge as it is discussed in the knowledge engineering community. We also talk about individual versus collective knowledge to conclude by showing how we consider daily knowledge and challenges to deal with in order to manage daily knowledge.
1.1. Knowledge
The notion of knowledge has been defined since Antiquity. Plato, for instance, defined thought as the intellectual model of objects. Heraclite went toward the definition of the logos as a triangle which distinguished thought, from expression, from reality. Saussure in his course [SAU 83] defined the base of the semiotic: a representation of knowledge embedded in an activity is related to a specific symbol. Currently, these representations are increasingly used to enhance learning from expertise and past experience. So, a human has to recognize concepts in the reference to make sense.
A sense is the combination of a signifier (the form which the sign takes) and a signified (the concept it represents). Within this theory, humans identify a sign from both the signifier and the signified. The semiotic triangle completed the representation of this theory with the use of three dimensions of knowledge: “sense”, “referee” and “symbol”. A human gives a sense to a symbol based on his/her referee (Figure 1.1).
Figure 1.1. Semiotic triangle
The opening of computer science to cognitive psychology in the 1950s, especially launched by the conference at Dartmouth in 1956, promoted the first analysis of how to represent human knowledge in a computational way. The first artificial intelligence studies concerned the development of an expert system (for instance the MYCIN system in the 1970s), in which expert knowledge is represented. The notion of the expert system became knowledge-based systems in the 1990s. A number of researchers studied how to represent knowledge based on logic. Thus, semantic networks and frames are defined with this aim [BRA 92]. Conceptual languages are also defined based on these theories. We mainly note conceptual graphs [SOW 14] and conceptual modeling language [SCH 94]. These studies are the basic principles of the current knowledge engineering theories in from which several techniques and notions took root: “conceptual models” of an expertise [BRE 94, AUS 94] and “ontologies” [FEN 01, BAC 00, CHA 04, KAS 02, KAS 05, GUA 98]. In these types of theories, knowledge is extracted from expert documents and by interviewing experts and represented in a conceptual model. This conceptual model can then be implemented using logics. Some methods such as MASK [ERM 00] use the conceptual representation to enhance learning between actors in an organization. Ermine in his method uses schematic forms in order to show links between concepts. He mentions the knowledge system and adds the representation of the context (borrowed from systemic science) to show different views of knowledge (Figure 1.2). His methods are largely used not in knowledge engineering but in knowledge management.
Figure 1.2. MASK views to represent knowledge. For a color version of the figure, see www.iste.co.uk/matta/knowledge.zip
The notion of knowledge management on the other hand began in management science in the 1990s [GRU 00, NON 95]. Knowledge management is the challenge of promoting the valorization of knowledge in an organization as a product. Some work in management science and economies goes beyond this and declares knowledge as the corner stone of a company [EVA 13, POL 66]. Polyani and Nonaka and Takeushi mentioned the notion of explicit and tacit knowledge. Nonaka and Takeushi defined the principle of transformation of knowledge between tacit and explicit knowledge (Figure 1.3).
Figure 1.3. SECI model [NON 95] . For a color version of the figure, see www.iste.co.uk/matta/knowledge.zip
The challenge of this type of work is how to support knowledge transformation in a company. So, we observe a number of techniques for this aim, for instance, the community of practices [LEV 97] and knowledge capitalization [DIE 02] approaches. The notion of corporate memory is defined as explicit and disembodied knowledge in a company [DIE 02]. Several techniques are inherited from knowledge engineering in order to enhance knowledge extraction in a company. These techniques have been adapted and completed. We note especially the MASK [ERM 00, MAT 02] and REX methods [MAL 93].
In management science, knowledge is considered as the production of interaction between actors [GRU 00]. To tackle this knowledge, techniques have to enhance this interaction. Grundstein [GRU 00] maintains that a system, which allows us to expertise identification, is better than knowledge extraction. Based on this principle, techniques allowing a knowledge map in a company have been presented [GRU 00, ERM 06, MAH 05].
If we refer to the definition of knowledge as the interaction between sign, reference and sense, a knowledge system must allow these three dimensions to be represented. In a company, actors produce knowledge because they interact continuously with these dimensions that are present in their work environment (interaction with a problem, interaction with other actors, interaction with a situation, etc.). So, managing knowledge in a company leads us to support these interactions. To this aim, some authors maintain that context will be represented with “How” an activity can be done and “What” type of concepts are used [COL 98]. Even current work in ontologies [CHA 04, KAS 02, KAS 05] promotes links between concepts and the documents from which they are produced. A lot of work presents tools enhancing document tagging [BEN 01, BEN 09] and concept construction [ZAC 07] in order to keep links between concepts and context.
In this book, we deal with knowledge as the interaction between an actor and his/her work environment. We study this interaction in cooperative activity in daily work. So, our main goal is to define techniques that help to enhance collaborative knowledge.
1.2. Daily knowledge
Daily knowledge consists mainly of know-how produced in daily work by a human. In the study by Richard [RIC 90], daily knowledge is considered as episodic memory, which contributes to build epistemic knowledge (or deep knowledge, as we mention in knowledge engineering). So, daily knowledge is dependent on the context in which it is produced (activity, environments, tools, etc.). Representing this type of knowledge also leads to representing its context and, especially, the organization and the environment in which it is produced.
Related to this postulate, the generation of a sense as it is represented in the semiotic triangle (Figure 1.1) cannot be done without the recognition of the context, which led to producing the reference. This postulate is further verified when we believe that knowledge is produced by the interaction of an actor with his/her environment. So the challenge is how to capture the context of the production of knowledge and how to represent it in order to enhance the generation of sense when learning from this knowledge (Figure 1.4). “The learning content is context specific, and it implies discovery of what is to be done when and how according to the specific organizations routines” [EAS 07].
Figure 1.4. Enhancing daily knowledge
So the main challenge is how to manage daily knowledge? How can we keep track of it by considering all the elements of the environment that contribute to its production: interaction, organization, roles, tasks, constraints, rules, means, methods, goals, products, artifacts, etc.
Currently, in organizations, collaborative activities are becoming more and more present. Dealing with the complexity of problems, actors have to solve problems in a collaborative way by interacting with other actors. So, we believe that observing daily knowledge production leads to dealing with collaborative activities. In this book, we study knowledge produced in collaborative activities, which we call “collaborative knowledge”.
1.3. Individual versus collaborative knowledge
As noted above, we deal with knowledge as the interaction between an actor and his/her environment. Some approaches in knowledge management study how to represent individual knowledge while others help to enhance collaborative knowledge. Before detailing our study on the management of collaborative knowledge from daily work, let us discuss the difference between individual and collaborative knowledge (Figure 1.5).
Figure 1.5. Difference between individual and collaborative knowledge
1.3.1. Difference in the nature of captured knowledge
In fact, knowledge capitalized with knowledge engineering approaches is related to experience. This experience is built along the activities of an expert in which a lot of experiments are analyzed and structured by the expert; knowledge engineering approaches are based on the cognitive psychology theory that enunciate that a human develops a mental schema and routines when repeating activities. In knowledge engineering, approaches tend to explain this mental schema by showing heuristic rules in different forms: the “what” manipulated the “why” of a behavior and the “how” of activities. Strategies and routines are so represented at what we call the conceptual level [AUS 94, BRE 94]. Newell [NEW 82] calls this level the knowledge level, in which behavior laws and rational actions must be represented. Observation of individual activity is not sufficient; knowledge engineering approaches developed several techniques to interact continuously with the expert in order to explain strategies and routines.
Note that knowledge observed in a collaborative activity is related to one experiment, i.e. a project. Actors in a company move continuously and collaborate in different projects. Each of them builds his/her own individual knowledge in his/her field. So, knowledge observed in a collaborative activity is related to episodic memory [RIC 90]. We know that semantic memory is built by repeating activities and aggregating information and data. So, observations of several activities are needed to capture and structure knowledge in collaborative work. Figure 1.6 explains knowledge development in individual versus collaborative activities.
Figure 1.6. Knowledge produced in collaborative activities
1.3.2. Difference in the dimension to be considered
The professional memory contains knowledge from a field. Collaborative knowledge belongs to several fields. In fact, in daily work, several teams (of several companies) and several disciplines collaborate to carry out an activity. So there is a collaborative and organizational dimension to consider in collaborative knowledge. Profession knowledge is generally about problem-solving in a domain [CAS 05], whereas collaborative knowledge is about organization, negotiation and cooperative decision-making in a project. So, to study the representation of collaborative activity, representations such as tasks to do, strategies followed and concepts manipulated are not sufficient. We need other theories from cooperative and organization sciences to define a representation structure of this knowledge. In fact, we must represent knowledge about:
- 1) the organization of the activity: actors, skills, roles;
- 2) the process of the activity: tasks, resources;
- 3) collaborative problem-solving: propositions, argumentation, conflicts, negotiation;
- 4) the context: directives, rules, techniques, constraints.
1.3.3. Difference in capturing of knowledge
The realization of a project within a company involves several actors from sometimes different groups and companies. For example, in concurrent engineering [SOH 92], several teams from several companies and several disciplines collaborate to carry out a project (Figure 1.7). The several teams are regarded as co-partners who share the decision-making during the realization of the project. This type of organization is generally dissolved at the end of the project. In this type of organization, the knowledge produced during the project’s realization has a collaborative dimension, which is generally volatile. The documents produced in a collaborative activity are not sufficient to keep track of this knowledge, which even the head of the project cannot explain. This dynamic characteristic of knowledge is due to cooperative problem-solving where various ideas are considered to build a solution. So extraction knowledge by interviewing experts or from documents as suggested in knowledge engineering approaches is not sufficient to show different aspects of the projects and especially negotiation [BEK 03]. Traceability and continuous knowledge capturing are needed to extract knowledge from collaborative activities.
Figure 1.7. Concurrent engineering activity
1.4. Challenge to manage daily knowledge
Daily knowledge can then be defined as knowledge observed in daily work by the interaction of actors with their environment in a company. So how can we continually store this knowledge and represent it in order to enhance learning in an organization?
In knowledge management, several researchers have developed principles and techniques to enhance learning in an organization. We can note the following two categories based on Nonaka and Takeushi’s [NON 95] SECI theory:
- 1) Explaining knowledge and developing a memory. In this work, we can note the process by Dieng [DIE 02] in which steps such as explanation, storing, updating and developing knowledge in a corporate memory are recommended. Ermine in his MASK method [ERM 00] proposes techniques to support these four steps in order to capitalize knowledge and develop a knowledge book as corporate memory. In this method, knowledge engineers co-build the knowledge book directly with domain experts, as there is an important interaction with the expert team in order to enhance learning and evolution of the memory. Other knowledge engineering approaches have been adapted as Common KADS [FEN 01] to support knowledge management. We note especially, the use of ontologies [GUA 98, FEN 01, KAS 02] to handle the semantic web as a cognitive documents memory in an organization. Techniques such as text mining [OTT 07, CHA 04, BAC 00] are then used in order to help in the definition of an ontology in a domain. Concepts in this type of ontology are linked to documents in order to represent their context. So, a number of works on tagging documents are then developed for this aim. We note especially the Porfyry system [BEN 10] (Figure 1.8), which helps to tag not only a piece of text but also images, and link them to concepts. Cahier [ZAC 07, BEN 10] developed the Agoare system that helps in ontology co-building in a cooperative activity (Figure 1.9). So, the system proposes a multiview representation of concepts and a vote workflow in order to co-define concepts in what he calls a “hypertopic tree”. Agoare helps actors to annotate and tag pieces of documents as a link to the topic identified.
- 2) Toward socialization and promoting interaction between actors. The basic principles in this type of approach are to define techniques in order to enhance interaction as a knowledge transfer between actors in an organization. Gundstein [GRU 00] proposes in the Gameth model to identify knowledge stakeholders and to represent a knowledge map that points out the knowledge type and stakeholders in an organization (Figure 1.10). He proposes techniques to identify knowledge types by studying process flows in an organization. Ermine [ERM 06] joins Grundstein by proposing techniques to define a knowledge map by exploring domains and services in a company. He also proposes a number of criteria in order to characterize knowledge, for instance, as strategic, rare, etc. (Figure 1.11). Based on this work on a knowledge map, some work tries to build data mining techniques in order to develop these maps in a company [BRA 13]. Other works develop supports to help actors to share their practices using discussion forums for instance. We can note community of practice studies [LEV 97]. In these types of techniques, discussion between actors having a given practice are stimulated by questions about activities and problems. The success of a community of practices depends mainly on their dynamic animators. Community of practices can be seen as a precursor to the current thematic social networks.
Figure 1.8. Porphyry system: tagging images and text [BEN 10]
Sharing documents, information and experiments without structuring this information and feedback analysis as used currently on community of practices, social network and information sharing systems as support of knowledge sharing is not sufficient to enhance learning. In fact, the “how” is shared but not the “what” and the “why” of activities. Behavior laws provide strong semantics to emphasize the reason behind this behavior, ready to be reproduced to solve new problems. The knowledge level by Newell [NEW 82] is necessary to promote learning in an organization. This principle is also the basis of pedagogical science, in which the law and the principle of a fact must be explained and then illustrated through examples. A learner has to define strategies and law as his knowledge ready to be applied to solve problems and to deal with new situations. So the challenge is not only how to capture knowledge from daily activity but how to structure it in behavior laws and strategies.
Figure 1.9. Agorae system principle [CAH 04]. For a color version of the figure, see www.iste.co.uk/matta/knowledge.zip
Figure 1.10. Gameth model [GRU 00]
Figure 1.11. Knowledge criteria [ERM 06]
Some techniques such as the REX method enhance the capturing and structuring of daily knowledge [MAL 93]. In this method, an actor has to complete an experience feedback form as a report of his/her daily activity. Each form is structured in order to show the definition of a problem and how the actor solves it [REN 08]. A lexicon that offers easy access to these forms indexes experience feedback forms. Forms can also be organized and indexed using different viewpoints in order to reflect the domain diversity in an organization.
Capturing and use of memory must be integrated in a work environment. Actors in a company have their own habits and it is very difficult to change their daily work to add another activity in order to capture knowledge. So the challenge is how to capture knowledge without disrupting activities? The failure of knowledge management approaches in organizations is mainly due to this problem. Due to concurrence, actors do not have enough time to reflect on their daily activities. They need to be stimulated and interrogated to do that. My experience on using the MASK method proves that experts do not have enough time to take 1 h from their time to only validate interview results. Note also that this validation is not done frequently.
To enhance learning from daily knowledge, first information must be structured in order to emphasize behavior laws and then examples must illustrate these laws. The knowledge production context must be represented in order to help in knowledge recognition and use. Organization actors need concrete examples in order to understand deep knowledge and learn from it. Otherwise, in a cooperative activity several domains are involved, so different viewpoints must be considered to support access to knowledge.
1.5. Conclusions
To sum up, the challenge to manage daily knowledge is to deal with the following:
- – how to capture information and interaction from daily activities without perturbing actors?
- – how to structure the information captured in order to explain the deep knowledge and behavior laws?
- – how to implement learning techniques from knowledge in daily work?
In this book, we deal with cooperative activities. As noted above, we believe that currently the main activities are realized in collaboration with actors in an organization. We then observe interaction between actors and between actors and their environment as a source of knowledge production in an organization.
1.6. Bibliography
[AUS 94] AUSSENAC-GILLES N., MATTA N., “Making a method of problem solving explicit with MACAO”, International Journal of Human-Computer Studies, vol. 40, pp. 193–219, 1994.
[BAC 00] BACHIMONT B., “Engagement sémantique et engagement ontologique: conception et réalisation d’ ontologies en ingénierie des connaissances”, Ingénierie des connaissances: evolutions récentes et nouveaux défis, pp. 305–323, 2000.
[BEK 03] BEKHTI S., MATTA N., “A formal approach to model and reuse the project memory”, Journal of Universal Computer Science, available at http://www.jucs.org/, vol. 6, pp. 12–22, 2003.
[BEN 01] BENEL A., EGYED-ZSIGMOND E., PRIE Y.,et al., “Truth in the digital library: fromontological to hermeneutical systems”, Research and Advanced Technology for Digital Libraries, Springer, Berlin/Heidelberg, pp. 366–377, 2001.
[BEN 09] BENEL A., LEJEUNE C., “Partager des corpus et leurs analyses à l'heure du Web 2.0”, Degrés: Revue de Synthèses à Orientation Sémiologique, vol. 36, pp. 136–137, 2009.
[BEN 10] BENEL A., ZHOU C., CAHIER, J.P., “Beyond Web 2.0. and beyond the Semantic Web”, From CSCW to Web 2.0: European Developments in Collaborative Design, Springer, London, pp. 155–171, 2010.
[BRA 92] BRACHMAN RONALD J., HECTOR J., “Levesque, and Raymond Reiter”, Knowledge Representation, MIT press, 1992.
[BRA 13] BRAHAMI M., ATMANI B., MATTA N., “Knowledge mapping evolution guided by data mining”, International Journal of Computer Applications, vol. 72, pp. 1–30, 2013.
[BRE 94] BREUKER J., VAN DE VELDE W. (eds.), Common KADS Library for Expertise Modelling: Reusable Problem Solving Components, IOS Press, vol. 21, 1994.
[CAH 04] CAHIER J.P., ZACKLAD M., MONCEAUX A., “Une application du web socio-sémantique à la définition d’un annuaire métier en ingénierie”, 15èmes Journées francophones d’Ingénierie des Connaissances, Presses Universitaires de Grenoble, pp. 29–40, 2004.
[CAS 05] CASTILLO O., MATTA N., “Definition of a practical learning system”, International Conference on Information Technology Based Higher Education and Training ITHET, Saint Domingue, pp. 14, 7–9 July 2005.
[CHA 04] CHARLET J., BACHIMONT B., TRONCY R., “Ontologies pour le web sémantique”, Revue I3, p. 31, 2004.
[COL 98] COLIN E., SPENDER J.C., “Managerial and organizational cognition”, Theory Methods and Research, SAGE, Paris, 1998.
[DIE 02] DIENG-KUNTZ R., CORBY O., GANDON F. et al., Méthodes et outils pour la gestion des connaissances, 2nd ed., Dunod, 2002.
[EAS 07] EASTERBY-SMITH M., LYLES M.A., Handbook of Organizational Learning and Knowledge Management, Blackwell, 2007.
[ERM 00] ERMINE J.L., La gestion de connaissances, Hermès Science, 2000.
[ERM 06] ERMINE J.L., BOUGHZALA I., TOUNKARA T., vol. 21, “Critical knowledge map as a decision tool for knowledge transfer actions”, Electronic Journal of Knowledge Management, vol 4, no. 2, pp. 129–140, 2006.
[EVA 13] EVANS J.S.B., OVER D.E., Rationality and Reasoning, Psychology Press, 2013.
[FEN 01] FENSEL D., Ontologies, Springer, Berlin/Heidelberg, pp. 11–18, 2001.
[GRU 00] GRUNDSTEIN M., “From capitalizing on company’s knowledge to knowledge management”, Knowledge Management, Classic and Contemporary Works, vol. 12, pp. 261–287, 2000.
[GUA 98] GUARINO N., “Formal ontology in information systems”, Proceedings of the First International Conference (FOIS’98), IOS Press, Trento, vol. 46, Italy,6–8 June 1998.
[KAS 02] KASSEL G., “OntoSpec: une méthode de spécification semi-informelle d’ontologies”, Actes des 13e journées francophones d’Ingénierie des Connaissances, pp. 75–87, 2002.
[KAS 05] KASSEL G., Integration of the DOLCE top-level ontology into the OntoSpec methodology, LaRIA Research Report 2005-08, 2005, available at: http://hal.ccsd.cnrs.fr/ccsd-00012203
[LEV 97] LEVINE M., PERKINS D.V., LEVINE, M., Principles of Community Psychology, New York, NY, Oxford University Press, 1997.
[MAH 05] MAHE S., “PUMEO: un modèle actif pour la gestion des connaissances tacites et expticites dans l’entreprise”, Ingénierie des connaissances, Editions L'Harmattan, p. 321, Paris, 2015.
[MAL 93] MALVACHE P., PRIEUR P., “Mastering corporate experience with the REX method”, Proceedings of ISMICK’93, Compiegne, October 1993.
[MAT 02] MATTA N., ERMINE J.L., AUBERTIN G. et al., “Knowledge capitalization with a knowledge engineering approach: the MASK method”, Knowledge Management and Organizational Memories, Springer US, pp. 17–28, 2002.
[NEW 82] NEWELL A., “The knowledge level”, Artificial Intelligence Journal, vol. 19, no. 2, 1982.
[NON 95] NONAKA I., TAKEUCHI H., The Knowledge-Creating Company: How Japanese Companies Create the Dynamics of Innovation, Oxford UniversityPress, 1995.
[POL 66] POLYANI M.,POLYANI K., The Great Transformation, Rinehart, New York,1966.
[REN 08] RENAUD J., MORELO B., FUSHS B. et al., “Retour d'expérience et connaissances industrielles”, Outils et démarches AFNOR, 2008.
[RIC 90] RICHARD J.F., Les activités mentales, Comprendre, raisonner, trouver des solutions, Armand Colin, Paris, 1990.
[SCH 94] SCHREIBER G., WIELINGA B., AKKERMANS H. et al., “CML: The Common KADS conceptual modeling language”, A Future for Knowledge Acquisition, Springer, Berlin/ Heidelberg, pp. 1–25, 1994.
[SOH 92] SOHLENIUS G., “Concurrent engineering”, Annals of the CIRP, vol. 41, no. 2, pp. 645--655, 1992.
[SOW 14] SOWA JOHN F., ed., Principles of Semantic Networks: Explorations in the representation of knowledge, Morgan Kaufmann, 2014.
[ZAC 07] ZACKLAD M., CAHIER J. P., BENEL A., et al., “Hypertopic: une métasémiotique et un protocole pour le Web socio-sémantique”, 18ème journées francophones d'ingénierie des connaissances (IC2007), pp. 217–228, July 2007.
Chapter written by Nada MATTA.