C Porcel et al Expert Systems with Applications 36(2009)5173-518 information, in order to allow a higher flexibility in the processes 3.3. Activity of siRE2IN of communication of the system. Therefore the system uses differ- ent label sets(S1, S2,..)to represent the different concepts to be The system activity can be described briefly in three steps assessed in its filtering activity. These label sets Si are chosen from those label sets that compose a LH, 1.e S E LH. We should point An expert receives or finds information about a research out that the number of different label sets that we can use is lim resource and inserts it into the system. ed by the number of levels of LH, and therefore, in many cases the Then, the system runs the matching process to determinate the label sets S and S, can be associated to a same label set of LH but fitting users to receive the information and send them an email with different interpretations depending on the concept to be with the information, the calculated relevance degree and modeled In our system, we distinguish between three concepts ommendations about possible collaborations with other users. that be assessed Once the users have received the information, they can change the kind of information they want to receive in the future, by portance degree(si)of a UNESCO code with respect to a resource updating their user profiles cope or user preferences. relevance degree (Si)of a resource for a researcher or for a In Fig. 4 we can see the main page of the system once the user is logged. Users can access the different options depending on the ompatibility degree (S3)between a researcher and a company, permissions they have assigned (user, expert or administrator ). between researchers of diferent groups and between different 3.3.1. Users insertion process This process consists in to incorporate users'data into the Following the linguistic hierarchy shown in Fig. 2, in our system system. It presents a form where the users insert their personal we use the level 2 (5 labels)to assign importance degree(S1=S) information, collaboration preferences and preferences about the and the level 3(9 labels)to assign relevance degrees(Sz d resources. Users are invited to define their topics of interest and in e, atibility degrees(S3=S). Using this LH the linguistic terms choose importance degrees(assessed in S,)associated with ach level are In order to gather information about users we use a hybrid ap- S=( bo=Null=N, b1=Low=L, b2=Medium=M, b3=High= proach, that is, when we insert a new user we use implicit inforn tion to generate the profile and afterwards the users can update S=co=Null=N, i=Very Low=VL, C2=Low=L, 3=More. their profiles following the explicit approach. Initially a user has Less low=MLL. C4=Medium=M, Cs=More Less High=MLH, associated the topics of interest of his/her research group or com- C6=High=H, c,=Very High= VH Cg=Total=T pany, but he/she can modify them. Each group signed one or more UNESCO codes, so the system assigns him/her So Therefore, for a resource i we have a vector representing its the UNESCO codes of level 2 of his/her group or company with importance degree Total((b4, 0), withb4 E S1). The other positions have a value Null((bo, 0), withbo E S1). Later, the users can update their profiles always they want, accessing to the system and edit where each component VR DE SI, with j 48, stores a lin- ing the uneSco codes or the importance degrees which they have guistic 2-tuple indicating the importance degree of the UNESCo assigned code j with regard to the resource i. These 2-tuples are assigned The system registers users and assigns them an identifier by the tto technician (email)and a password Finally, the users receive a confirmation To represent the topics of interest in the user profiles we fol email with the inserted information he same method, using a vector VU for each user of the sy Then, for the user x, we have a vector Example 1. In this example, we are going to insert a new user. The VUx=(VUx[1]. VUx2],., VUx[248) user fills the form with his her information. Let us suppose the user belongs to a group which works in Science of Nutriment, then he/ where each component VUxD E SI, with y=1... 248, stor she has assigned the UNESCO code 3206. Then, to define the vector ustic 2-tuple indicating the importance degree of the of topics of interest the system assigns the user this code(3206) code y with regard to the preferences of the user x. These with degree Total(b4, 0), witha E S1). With this information the are also assigned by the experts, but the users can edit them when user profile is represented by a vector of topics of interest with the they want. Recommender System about Research Resources user managemen Fig 4. Main page of SIRE2IN.information, in order to allow a higher flexibility in the processes of communication of the system. Therefore the system uses differ￾ent label sets ðS1; S2; ...Þ to represent the different concepts to be assessed in its filtering activity. These label sets Si are chosen from those label sets that compose a LH, i.e., Si 2 LH. We should point out that the number of different label sets that we can use is lim￾ited by the number of levels of LH, and therefore, in many cases the label sets Si and Sj can be associated to a same label set of LH but with different interpretations depending on the concept to be modeled. In our system, we distinguish between three concepts that can be assessed: importance degree (S1) of a UNESCO code with respect to a resource scope or user preferences, relevance degree (S2) of a resource for a researcher or for a company, compatibility degree (S3) between a researcher and a company, between researchers of different groups and between different companies. Following the linguistic hierarchy shown in Fig. 2, in our system we use the level 2 (5 labels) to assign importance degree (S1 ¼ S 5 ) and the level 3 (9 labels) to assign relevance degrees (S2 ¼ S9 ) and compatibility degrees (S3 ¼ S9 ). Using this LH the linguistic terms in each level are S5 ¼ fb0 ¼ Null ¼ N;b1 ¼ Low ¼ L;b2 ¼ Medium ¼ M;b3 ¼ High ¼ H;b4 ¼ Total ¼ Tg S9 ¼ fc0 ¼ Null ¼ N; c1 ¼ Very Low ¼ VL; c2 ¼ Low ¼ L; c3 ¼ More Less Low ¼ MLL; c4 ¼ Medium ¼ M; c5 ¼ More Less High ¼ MLH; c6 ¼ High ¼ H; c7 ¼ Very High ¼ VH; c8 ¼ Total ¼ Tg Therefore, for a resource i we have a vector representing its scope: VRi ¼ ðVRi½1; VRi½2; ... ; VRi½248Þ; where each component VRi½j 2 S1, with j ¼ 1; ... ; 248, stores a lin￾guistic 2-tuple indicating the importance degree of the UNESCO code j with regard to the resource i. These 2-tuples are assigned by the TTO technicians. To represent the topics of interest in the user profiles we follow the same method, using a vector VU for each user of the system. Then, for the user x, we have a vector: VUx ¼ ðVUx½1; VUx½2; ... ; VUx½248Þ; where each component VUx½y 2 S1, with y ¼ 1 ... 248, stores a lin￾guistic 2-tuple indicating the importance degree of the UNESCO code y with regard to the preferences of the user x. These 2-tuples are also assigned by the experts, but the users can edit them when they want. 3.3. Activity of SIRE2IN The system activity can be described briefly in three steps: An expert receives or finds information about a research resource and inserts it into the system. Then, the system runs the matching process to determinate the fitting users to receive the information and send them an email with the information, the calculated relevance degree and rec￾ommendations about possible collaborations with other users. Once the users have received the information, they can change the kind of information they want to receive in the future, by updating their user profiles. In Fig. 4 we can see the main page of the system once the user is logged. Users can access the different options depending on the permissions they have assigned (user, expert or administrator). 3.3.1. Users insertion process This process consists in to incorporate users’ data into the system. It presents a form where the users insert their personal information, collaboration preferences and preferences about the resources. Users are invited to define their topics of interest and choose importance degrees (assessed in S1) associated with them. In order to gather information about users we use a hybrid ap￾proach, that is, when we insert a new user we use implicit informa￾tion to generate the profile and afterwards the users can update their profiles following the explicit approach. Initially a user has associated the topics of interest of his/her research group or com￾pany, but he/she can modify them. Each group or company has as￾signed one or more UNESCO codes, so the system assigns him/her the UNESCO codes of level 2 of his/her group or company with importance degree Total (ðb4; 0Þ; withb4 2 S1). The other positions have a value Null (ðb0; 0Þ; withb0 2 S1). Later, the users can update their profiles always they want, accessing to the system and edit￾ing the UNESCO codes or the importance degrees which they have assigned. The system registers users and assigns them an identifier (email) and a password. Finally, the users receive a confirmation email with the inserted information. Example 1. In this example, we are going to insert a new user. The user fills the form with his/her information. Let us suppose the user belongs to a group which works in Science of Nutriment, then he/ she has assigned the UNESCO code 3206. Then, to define the vector of topics of interest the system assigns the user this code (3206) with degree Total (ðb4; 0Þ; withb4 2 S1). With this information the user profile is represented by a vector of topics of interest with the following values: Fig. 4. Main page of SIRE2IN. 5178 C. Porcel et al. / Expert Systems with Applications 36 (2009) 5173–5183