Decision Support System of Lecturer Selection Recommendation with Collaborative Filtering Method

International Journal on A d v a n c e d S c i e n c e E n g i n e e r i n g Information Technology

Vol.10 (2020) No. 2 ISSN: 2088-5334

Decision Support System of Lecturer Selection Recommendation with Collaborative Filtering Method

A n t o n S e t i a w a n H o n g g o w i b o w o a , H e r o W i n t o l o a, Y u l i a n i I n d r i a n i n g s i h a, R e g i n a M a u l a A d i b a a a Department o f Informatics, Adisutjipto College O f Technology, Janti Street, Blok-R, Lanud Adisutjipto Yogyakarta

E-mail: anton_s_h@yahoo.com

Abstract

— The use of decision support systems for the selection of the best lecturers a t A disutjipto Institute of Technology (STTA) is not yet at the level of application. The average te rtiary institution in selecting the best lecturers uses certain criteria, for example the criteria in teaching, research and com m unity service. O f the three indicators of lecturer perform ance in Indonesia, we only use two indicators, nam ely research and com m unity service. F rom these two indicators, we m ade five criteria, nam ely the num ber of presenters a t the conference, the am ount of com m unity service, the num ber of unpublished research, the num ber of published research, and the num ber of citations of scientific articles. The selection was m ade by users, nam ely STTA Director, Vice D irector for Academic Affairs, Vice D irector for Financial Affairs and Com m unity Service Research C enter to 62 lecturers who w ere active in research, com m unity service and publications indexed on Google Scholar. This user restriction is adjusted to the organizational stru ctu re a t o ur university, w here the five users have authority in assessing the perform ance of lecturers and giving aw ards to lecturers who are declared as the best lecturers. A fter C ollaborative Filtering m ethod is done to predict the final result using the rating given by the user. Based on the results of the system testing, it was concluded th a t the system th a t was built could be a solution to help select lecturers who w ere eligible to be given aw ards as the best of lecturer in the field of research and com m unity service.

Keywords

— decision support system; lecturer recom m endations; collaborative filtering m ethod.

I. In t r o d u c t io n

D ecision support system s evolve along w ith technological developm ents that support com puterization in d ecision­

m aking. A lm ost every sm all and large com pany has a decision support system , including universities. A n exam ple is decision support system s for the selection of L ecturer in the field of research. L ectu rer at the A disutjipto C ollege of T echnology (STTA) Y ogyakarta every y ear is alw ays selected through selection to be aw arded as an outstanding lecturer. The aw arding of outstanding lecturers is expected to im prove the perform ance and dedication of lecturer and can help to im prove the value of accreditation for universities.

The use of decision support system in various fields has helped people to m ake various decisions, m ake decisions in agriculture for the diagnosis o f rice plants [1], com puter m aintenance, [2] and spraying w eeds on plants [3]. D ecision support system s can also be applied to a university related to scholarship provision using T O P S IS and W eighted Product m ethods [4], increasing lecturer satisfaction in term s o f preferences and ratings [5], and can im prove library perform ance efficiency in a collage in providing book recom m endations alternative to visitors w ith collaborative

filtering m ethods [6]. So that in the article m ade from research using collaborative filtering m ethods to support of the decision o f the choice o f lecturers w ith achievem ents in the field research.

II. Ma t e r ia la n d Me t h o d

A. Recommendation System

R ecom m endation system s are interm ediary program s or representatives that intelligently com pile list o f inform ation needed and m atch based on the w ishes o f users [7]. The recom m endation system s aim to suggest item s to users. The recom m endation system s directly advise users to item s that can m eet their needs and desires by narrow ing dow n inform ation in large database [8].

B. Collaborative Filtering Methods

C ollaborative filtering is the process filtering or evaluating item s using the opinions o f others [9]. A lthough the term collaborative filtering has only existed for about a decade, collaborative filtering derives from som ething that hum ans have done for centuries to share opinions w ith others.

C ollaborative filtering is one o f the algorithm s used to develop the recom m ender system and has proven to provide

excellent results. R ating is the m ost im portant elem ent of this algorithm ; rating is obtained from m ost users w here the user explicitly gives an assessm ent o f the product. The conclusions is the system gives reciprocity to the user by processing these data, as an illustration o f a scale o f 0 to 5 w hich indicates the m ost unpopular to the m ost preferred assessm ent according to the u s e r’s point o f view , this data allow s for statistical calculations w hich results indicate w hich product given a high rating b y user.

TABLE I Matrix Rating Table

s i m ( i , j )

=

TiuEU(Ru,i Ri)(Ru,j

^j)

i1 i2 i3 i4 in

U, 1 3

U2 5 4

U3 5 3

U4 4

Un

C ollaborative filtering uses a database obtained from the user. T here are tw o m ain com ponents in this data in order to m ake predictions for the recom m ender system , nam ely the user and the item , w hich in this case is a criterion. B o th from m atrix ratings in the form o f m user {u1, u2, u 3 ... um } and a list o f n item {i1, i2, i3 ... in}. W here each user is gives an assessm ent of the item in the form o f a rating on a scale o f 1 to 5. Iu1 denotes this rating. N ot all users give ratings to each product due to various factors, this cause the num ber of m issing values that result in data sparsely. U ser rating m atrix item s can be described w ith the ta b le1.

T here are two m ain approaches in the collaborative filtering m ethod, nam ely:

1) User-based collaborative filtering:

T his algorithm w orks based on the assum ption that each user is part of group that has sim ilarities w ith other users. T he basis o f the recom m endation w ith this algorithm is that the recom m endations produced the arranged based on item s that are liked by each user. R ecom m ended item s are the result of recom m endations according to w hat other users like. B ased on item s that have been chosen by the closest neighbor o f a user, item s that are likely to be chosen b y the user in the future are predicted [10]. A lgorithm s that are often used include the Pearson C o rrelation C oefficient (PCC) algorithm , the V ector S pace S im ilarity (VSS) algorithm .

2) Item-based collaborative filtering:

T his algorithm w orks to find relationships betw een item s based on the rating table to form a recom m endation for an item to the user. To produce recom m endations, first a correlation m odel betw een item s is needed w ith the aim o f know ing the relationship betw een item s according to the rating obtained.

C o rrelation m odels can be done offline using various other techniques, such as association rule, classifications or clustering. To m ake a recom m endation system using the item -based collaborative filtering m ethod, there are three steps that m ust be done, nam ely:

• D eterm ine the user item -rating matrix.

• C alculate S im ilarity w ith the form ula adjusted cosine sim ilarity.

Y*uEU(Ru,i Ru) JEuEU(Ru,j Ru)

(1)

D escription:

s i m( i , j )

= S im ilarity value betw een item i and item j

u e u

Ru,i Ri R_uj Rj Ru

= U ser set that evaluates item i and item j

= u U ser rating u on item s i

= A verage value o f rating item i

= u U ser rating on item j

= A verage value o f rating item j

= A verage rating for user u

C alculate rating predictions w ith W eight Sum:

T,iEl(S[i,j)xR[u,j))

P ui = ■

SiEl|!(i,j)| (2)

D escription:

P(u, i)

= Prediction o f user rating for item i

i E I

= Set o f item sim ilar to item i

' ( i , j )

= V alue o f S im ilarity betw een item i on item j

Ru,j

= u user rating on item j

C. Decision Support Systems

D ecision support system (DSS) is defined as a system intended to support m anagerial decision m aking in certain situation. D ecision support system s are intended to be a tool for decision m akers to expand their capabilities, but not to replace their ju d g m e n ts [11]. A pplication o f decision support system s can consist o f several subsystem s, including data m anagem ent, m odel m anagem ent, user interface and know ledge based m anagem ent [12].

III. Res u l t sa n d Dis c u s s io n

A. Rating

In recom m endation system w ith the collaborative filtering m ethod, the role o f rating is v ery im portant to determ ine the final item w orthy to be recom m ended. In calculating the D SS recom m endations o f lecturers w ith achievem ents in the field o f research, the rating obtained by each lecturer depends on the num ber o f values possessed by each criterion. R ating w ill be done autom atically based on the w eight given to each user w ho gives an assessm ent.

T he first thing to be done before giving a rating is to determ ine the criteria; In this case, five criteria have been determ ined, nam ely:

P1 = N um ber o f presenter

P2 = N um ber o f com m unity services P3 = N um ber o f unpublished research P4 = N um ber o f publish research P5 = N um ber o f citations

TABLE II Rating Table

No. Column Name Type of D ata

1. 5 Very Good

2. 4 Good

3. 3 Enough

4. 2 Less

5. 1 Very Less

A fter determ ining the criteria, the second phase is R ating is given from values 1 to 5, details o f the rating table determ ining the suitability rating o f each alternative criteria. can be seen T able II.

TABLE III Rating Weighting Table

No User C riteria

Code

Rating

1 2 3 4 5

BB BA BB BA BB BA BB BA BB BA

1. Director STTA P1 0 0 1 2 3 4 5 6 7 >7

P2 0 1 2 3 4 5 6 8 9 >9

P3 0 1 2 3 4 5 6 8 9 >9

P4 0 1 2 3 4 5 6 8 9 >9

P5 0 5 6 10 11 20 21 30 31 >31

2. Vice Director for Academic Affair

P1 0 0 1 1 2 3 4 4 5 >5

P2 0 1 2 3 4 5 6 8 9 >9

P3 0 1 2 2 3 5 6 7 8 >8

P4 0 0 1 2 3 4 5 5 6 >6

P5 0 5 6 15 16 25 26 40 41 >41

3. Vice Director for Financial Affair

P1 0 0 1 1 2 3 4 5 6 >6

P2 0 2 3 4 5 6 7 7 8 >8

P3 0 1 2 3 4 5 6 7 8 >8

P4 0 1 2 4 5 6 7 8 9 >9

P5 0 10 11 20 21 30 31 40 41 >41

4. Community Service Research Center

P1 0 1 2 2 3 4 5 5 6 >6

P2 0 1 2 3 4 5 6 7 8 >8

P3 0 2 3 5 6 7 8 9 10 >10

P4 0 1 2 4 5 5 6 7 8 >9

P5 0 5 6 10 11 20 21 30 31 >31

Description: BB = Lower limit value BA = Upper limit value

The third is w eighting each rating based on the value of each criterion. U sers w ho have the authority to perform system calculations carry out w eighting. U sers w ho have authority include the D irector STTA , V ice D irector for A cadem ic A ffair, V ice D irector for F inancial A ffair and C om m unity S ervice R esearch Center. W eighting w ill vary according to the rating w eight given by each user. The detailed table o f w eights can be seen in T able III.

T able III show s the range o f values for each criterion. In the rating colum n, w e can see the w eights for the 1-5 rating.

The rating is given based on the w eight o f the predeterm ined value. F or exam ple, the details o f the w eight o f the D irector S T T A give w eight to the criteria P1 to P5 w ith the detailed rating criteria.

P1 is the am ount as a presenter. For rating 1has a range o f low er lim it 0 and upper lim it 0 (range 0-0). For rating 2 has a range o f low er bound 1 and upper lim it 2 (range 1-2). For rating 3 has a range o f low er lim it 3 and upper lim it 4 (range 3-4). For rating 4 has a range o f low er lim it 5 and upper lim it 6 (range 5-6). For rating 5 has a low er lim it range o f 7 and an upper lim it range o f < 7. For exam ple, suppose that the lecturer has been a presenter 2 tim es than the P1 rating that the lecturer gets from the D irector STT A is 2, because the rating o f 2 for P1 has a range o f 1-2.

P2 is the num ber o f com m unity service. For rating 1 has a range o f low er lim its 0 and upper lim it 1 (range 0-1). For rating 2 has a range o f low er lim it 2 and upper lim it 3 (range 2-3). T he rating 3 has a low er lim it range o f 4 and upper lim it o f 5 (range 4-5). For rating 4 has a range o f low er lim it 6 and upper lim it 8 (range 6-8). For rating 5 has a low er lim it range o f 9 and upper lim it range o f < 9. For exam ple, in 1

year the lecturer has done com m unity service 4 tim es, then the P2 rating that the lecturer gets from the D irector STT A is 3, because rating 3 for P2 has a range o f 4-5.

P3 is the num ber o f u npublished research. For rating 1 has a range o f low er lim it 0 and upper lim it 1 (range 0-1).For rating 2 has a range o f low er lim it 2 and upper lim it 3 (range 2-3). The rating 3 has a low er lim it range o f 4 and upper lim it o f 5 (range 4-5). For rating 4 has a range o f low er lim it 6 and upper lim it 8 (range 6-8). For rating 5 has a low er lim it range o f 9 and an upper lim it range o f < 9. For exam ple, suppose that in 1 year, there w ere 3 unpublished research, then P3 rating that the lecturer got from the D irector STT A w as 2, because the rating 2 for P3 had a range o f 2-3.

P4 is the num ber o f published research. For rating 1 has a range o f low er lim it 0 and upper lim it 1 (range 0-1). For rating 2 has a range o f low er lim it 2 and upper lim it 3 (range 2-3). For rating 3 has a low er lim it range o f 4 and upper lim it o f 5 (range 4-5). For rating 4 has a range o f low er lim it 6 and upper lim it 8 (range 6-8). For rating 5 has a low er lim it range o f 9 and an upper lim it range o f < 9. For exam ple, suppose that in 1 y ear there are 6 p u b l i s h e d r e s e a r c h , the P4 rating that the lecturer gets from the D irector S T T A is 4, because the 4 rating for P4 has a range o f (6-8).

P5 is citation. For rating 1 has a range o f low er lim it 0 and an upper lim it 5 (range 0-5). For rating 2 has a range o f low er lim it 6 and an upper lim it 10 (range 6-10). T he rating 3 has a low er lim it range o f 11 and an upper lim it o f 20 (range 11-20). For rating 4 has a range o f low er lim it 21 and an upper lim it o f 30 (range 21-30). For rating has a range of low er lim it 31 and an upper lim it range is 1 30. For exam ple, suppose that the num ber o f citations from lecturers research

is 22 then the P5 rating that the lecturer gets from the D irector S T T A is 4, because the 4 to P5 rating has a range of 21 - 30.

A fter w eighting, the fourth is to convert the num ber of values the lecturer has into a rating according to the w eight entered. D etails o f the value o f each criterion P1 to P5 that is ow ned by the lecturer can be seen in T able IV. The values in T able V w ill be converted into ratings according to the w eight given by each user.

TABLE IV Detail Value Amount Table

F rom the detailed num ber o f values in T able 4.3 show s the H ero W . L ecturer has a num ber o f criteria valu es P1=0, criteria P2=0, P3=0, P4=4, P5=14. For the num ber o f second lecturers and so on according w ith T able IV.

The results o f conversion o f values to rating can be seen in T able V. T his table show s the results o f the conversion of each lecturer from the criteria P1 to P5 criteria. To get a rating, the num bers in T able IV are com pared w ith the w eights in T able III. Each lecturer has four different conversion results; this is because four users w ho have different rating w eights carry out the assessm ent.

TABLE V Conversion Result Table

The results o f the conversion o f criteria based on the w eight given by the D irector STT A for H ero W . L ecturer are as follow s:

• T he num ber o f criteria value P1 = 0, then the criteria rating P1 = 1

• T he num ber o f criteria value P2 = 0, then the criteria rating P2 = 1

• T he num ber o f criteria value P3 = 0, then the criteria rating P3 = 1

• T he num ber o f criteria value P4 = 4, then the criteria rating P3 =

• T he num ber o f criteria value P5 = 14, then the criteria rating P5 = 3.

For the conversion results, the criteria for the second lecturer and so on are in accordance w ith table V. The criteria value w eights can be seen in T able III. The num ber o f criteria values P1 to P5 can be seen in T able IV. The fifth step is to find the average rating o f each lecturer, w hich then b ecom es the final rating. E ach lecturer w ill have four average values obtained from four users, nam ely D irector S TTA , V ice D irector for A cadem ic A ffair, V ice D irector for Financial A ffair, and C om m unity Service R esearch Center.

T he calculation o f the average rating o f each lecturer obtained from the D irector S T T A is as follow s:

• A verage lecturer rating o f H ero W intolo 1+1+1+3+3 ^ 0

=--- = 1 . 0 5

• A verage lecturer rating o f A nton S etiaw an H.:

1+1+1+4+5

= --- —

2.4

5

• A verage lecturer rating o f M ardiana Iraw ati:

1 + 1 + 1 + 1 + 1

= i

= 5

~

• A verage lecturer rating o f A stika A yuningtyas:

1 + 1 + 1 + /+ 1 . ,,

= ---—

1.b

5

• A verage lecturer rating o f H aruno Sajati:

1 + 1 + 1 + 3 + 3

= ---—

1.b

5

For the final rating details o f all users can be seen in T able VI.

TABLE VI Final Rating Details

T he sixth step is to find sim ilarity betw een lecturers w ith the follow ing calculations:

No User L ecturer

nam e

C riteria

P1 P2 P3 P4 P5

1. Director STTA

Hero W. 1 1 1 3 3

Anton S.H 1 1 1 4 5

Mardiana I 1 1 1 1 1

Astika A. 1 1 1 4 1

Haruno S. 1 1 1 4 1

2. Vice Director for Academic Affair

Hero W. 1 1 1 3 2

Anton S. H. 1 1 1 5 5

Mardiana I 1 1 1 1 1

Astika A. 1 1 1 5 1

Haruno S. 1 1 1 5 1

3. Vice Director for Financial Affair

Hero W. 1 1 1 2 2

Anton S. H. 1 1 1 3 5

Mardiana I 1 1 1 1 1

Astika A. 1 1 1 3 1

Haruno S. 1 1 1 3 1

4. Community Service Research Center

Hero W. 1 1 1 3 3

Anton S.H 1 1 1 4 5

Mardiana I. 1 1 1 1 1

Astika A. 1 1 1 4 1

Haruno S. 1 1 1 4 1

Final Rating No Lecturer

Name

Director STTA

Vice Director

for Academic

Affair

Vice Director

for Financial

Affair

Community Service Research

Center

1. Hero W. 1.8 1.6 1.4 1.6

2. Anton S.H.

2.4 2.6 2.2 2.4

3. Mardian a I.

1 1 1 1

4. Astika A.

1.6 1.8 1.4 1.6

5. Haruno S.

1.6 1.8 1.4 1.6

Average 1.68 1.76 1.48 1.64

No. L ecturer Name C riteria

P1 P2 P3 P4 P5

1. Hero W. 0 0 0 4 14

2. Anton S. H. 0 0 0 6 72

3. Mardiana I. 0 0 0 0 4

4. Astika A. 0 0 0 6 0

5. Haruno S. 0 0 0 6 4

s i m ( i , j ) —

W f r u

Ri)(Ruj Rj)

(3)

JXueu(Ru,i~Ru) jEuEu(Ru,j~Ru)

B ased on the calculation o f sim ilarity above, a sim ilarity value is chosen w hich has the highest value, nam ely 1. T hen after obtaining sim ilarity values, the seventh step is to find predictions using the sim ilarity value that has been determ ined w ith the follow ing details:

• P rediction D irector STTA Pred A = = 1.2

|1| Pred B = — - = 1.8

|1|

Pred C = = 2.8

|1|

Pred D = = 2.8

|1|

1 3 2 Pred E = —— = 3.2

|1|

• P rediction V ice D irector for A cadem ic A ffair 1 1 /

Pred A = - - - = 1.4

|1| Pred B = = 2.0

|1|

Pred C = = 2.8

|1|

Pred D = = 2.8

|1|

1 3 / Pred E = —— = 3.4

|1|

• P rediction V ice D irector for F inancial A ffair Pred A = m 5 = 1.0

|1|

1 1 / Pred B = — — = 1.4

|1| Pred C = = 2.6

|1|

1 2 / Pred D = —— = 2.4

|1| Pred E = - ■4 5= 3.0

|1|

• P rediction C om m unity Service R esearch Center 1 1 3

Pred A = — - = 1.2

|1| Pred B = = 1.6

|1|

1 2 / Pred C = — - = 2.4

|1| Pred D = = 2.6

|1| Pred E = 1— 3 5 = 3.0

|1|

A fter the prediction calculation is obtained, then the last one is sorting the results o f the calculation o f the predictions of the five lecturers from each user, nam ely the D irector STTA , V ice D irector for A cadem ic A ffair, V ice D irector for Financial A ffair, and C om m unity Service R esearch Center.

C alculations are sorted from the highest num ber to the low est num ber. The results o f the prediction sequence are the final result o f the recom m endation.

B. Comparison o f Manual Calculation and System Calculation

The results o f the m anual calculation o f the DSS recom m endation o f the outstanding lecturers in the research w ere determ ined from the final value o f the prediction calculation. D etails o f the final results o f the m anual calculation o f DSS recom m endation can be seen in T able VII.

TABLE VII

Detailsof Manual Calculation Result

No.

Final rating Director

STTA

Vice Director for Academic

Affair

Vice Director for

Financial Affair

Community Service Research

Center 1. Anton S. H.

(2.4)

Anton S. H.

(2.6)

Anton S. H.

(2.2)

Anton S. H.

(2.4) 2. Hero W. (1.8) Haruno S. (1.8) Haruno S.

(1.4)

Haruno S.

(16) 3. Haruno S. (1.6) Astika A. (1.8) Hero W.

(1.4)

Hero W.

(16) 4. Astika A. (1.6) Hero W. (1.6) Astika A.

(1.4)

Astika A.

(16) 5. Mardiana I. (1) Mardiana I. (1) Mardiana I.

(1)

Mardiana I.

(1) IV. Co n c l u s io n s

F rom the results o f testing on D ecision S upport System s R ecom m endations o f achieving lecturers in the field of research w ith collaborative filtering m ethods can be taken som e conclusions. The first is based on sam ple of data that has been tested that is as m any as 5 lecturer indicate that the results o f application calculations are the sam e as the m anual calculation results. The second conclusion is that this system can be used as a solution to provide recom m endations for lecturers who are eligible to be aw arded as outstanding lecturers in the field o f research.

Re f e r e n c e s

[1] A. S. Honggowibowo, “Sistem Pakar Diagnosa Penyakit Tanaman Padi Berbasis Web Dengan Forward Dan Backward Chaining,”TELKOMNIKA (Telecommunication Computing Electronics and Control), vol. 7, no. 3, p. 187, Dec. 2009.

[2] F. Sukmana and F. Rozi, “Decision Support System On Computer Maintenance Management SystemUsing Association Rule and Fisher Exact Test One Side P-Value,” TELKOMNIKA (Telecommunication Computing Electronics and Control), vol. 15, no. 4, p. 1841, Dec.

2017.

[3] R. M. Sampurno, K. Boro Seminar, and Y. Suharnoto, “Weed Control Decision Support System Based on Precision Agriculture Approach,” TELKOMNIKA (Telecommunication Computing Electronics and Control), vol. 12, no. 2, p. 475, Jun. 2014.

[4] S. ‘Uyun and I. Riadi, “A Fuzzy Topsis Multiple-Attribute Decision Making for Scholarship Selection,” TELKOMNIKA (Telecommunication Computing Electronics and Control), vol. 9, no.

1, p. 37, Apr. 2011.

[5] H. Babaei, J. Karimpour, and A. Hadidi, “Applying Hybrid Fuzzy Multi-Criteria Decision-Making Approach to Find the Best Ranking for the Soft Constraint Weights of Lecturers in UCTP,” International Journal of Fuzzy Systems, vol. 20, no. 1, pp. 62-77, Feb. 2017.

[6] P. R. Hasibuan, A. S. Honggowibowo, and H. Wintolo, “Sistem Pakar Rekomendasi Peminjaman Buku Di Perpustakaan Sekolah Tinggi Teknologi Adisutjipto Yogyakarta Dengan Metode Penyaringan Kolaboratif,” Compiler, vol. 1, no. 1, May 2012.

[7] N. Mittal, R. Nayak, M. C. Govil, and K. C. Jain, “Recommender System Framework Using Clustering and Collaborative Filtering,”

2010 3rd International Conference on Emerging Trends in Engineering and Technology, Nov. 2010.

[8] S. Parvatikar and B. Joshi, “Online book recommendation system by using collaborative filtering and association mining,” 2015 IEEE International Conference on Computational Intelligence and Computing Research (ICCIC), Dec. 2015

[9] J. B. Schafer, D. Frankowski, J. Herlocker, and S. Sen,

“Collaborative Filtering Recommender Systems,” Lecture Notes in Computer Science, pp. 291-324.

[10] M.S. Babu & B.R.S. Kumar, An implementation of the user-based collaborative filtering algorithm, International Journal of Computer Science and Information Technologies 2,3, 2011: 1283-86.

[11] E. Turban, J.E. Aronson, & T.P. Liang, Decision Support Systems and Intelligent Systems (Sistem Pendukung Keputusan dan Sistem Cerdas) edisi ketujuh jilid 1, Yogjakarta: Andi Offset, 2005.

[12] Kusrini, Concepts and Applications of Decision Support Systems, Andi Offset, 2007.

L E M B A R

H A S I L P E N I L A I A N SE JAVVAT S E B I D A N G A T A U P E E R R E V I E W K A R Y A I L M I A H : J U R N A L I L M I A H

Judul K a ry a Ilm iah (A rtikel) P en u lis Jurnal Ilm iah Id en titas Jurnal Ilm iah

D ecisio n su p p o rt system o f lecturer selection reco m m en d atio n w ith co llab o ra tiv e filtering m ethod A n to n Setiaw an H ongg o w ib o w o , H e r o W in t o l o , Y uliani Indrian in g sih . R eg in a M aula

A diba

a. N a m a Jurnal : IJA S E IT (In tern atio n a l Journal on A dvanced Science. E n g in eerin g and Inform ation T echnology) b. N o m o r/V o lu m e : 2 / 10

c. Edisi (b u lan /tah u n ) : 2020

d. P e n erb it : IN S IG H T - Ind o n esian S ociety for K n ow ledge and H um an D evelopm en

e. url d okum en :

h ttp ://in s ig h ts o d e t\ .o ra/o iascii m d cx .p h p ...ia scit/article.p ew 8407 t..', 1360 H asil Penilaian Peer Review :

Komponen Yang Dinilai

Nilai Millesimal Ju r n a l ilmiah Nilai Akhir Yang

Diperoleh Internasi

onal Bereput

asi H

Interna sional

□

Nasio nal Terak redita si

□

Nasiona 1 Tidak T erakre

ditasi n

Nasio nal Terin

deks DOAJ

□

J u r n a l Nasional Terakreditasi Remristekdikti Peringkat 1

n 2 n

3 n

4

□ 5

□ 6

□ a. K elengkapan unsur isi buku

(10%)

4

3 , L

b. R uang lingkup dan kedalam an pem bahasan (30%)

12

S,<f

c. K ecukupan dan kem utahiran data/inform asi dan m etodologi (30%)

12

d. K elengkapan unsur dan kualitas penerbit (30%)

12

Total

=

^{(100%) 40}

?i.fi

Kontribusi Pengusul (Penulis Kedua dari Tiga Penulis)

(13,3% x3l^ S — i <2

... )

Komentar Peer Review

1. Tgntan C<rv 2. T entar

StCh

3. Tentai d i « 4. T entar

6i\y

g kelengkapan unsyr isi buku ....

jkap W**/) jfan A A r /\vnr4,C

h k u s ....

tnsQ aw C Y fy w ftw fa /-

latAXs r a v \yn t$

g ruang Lingkup dan kedalam an pem bahasan

S#

ny\. J tv ijA .r' pm m h d iA ir, C

^{a k w v}

vev

g kecukupan dan kem utakhiran data/ijiform asi dan m etodologi.

'At..Aftvs... mhjddo^iyuyi'...MCVf Hiubik w v

g K elengkapan unsur danj<ualitas p e n e r b i t . .

...f^l/1 1 ... ...

Y ogyakarta.

- I I -

²⁰²¹

Review er I

v .

I

( S u l n r ' v * * ' ' , P n 0 ,

Unit Kerja:

L E MB AR

HA S I L P E N I L AI A N S E J A W A T S E B I D A N G AT AU P E E R R E V I E W K A R Y A I LMI AH : J U R N A L I LMIAH

Judul K arya Ilm iah (A rtikel) Penulis Jurnal Ilm iah Identitas Jurnal Ilm iah

Hasil Penilaian P eer R eview :

: Decision support system o f lecturer selection recom m endation with collaborative filtering nietho : Anton Setiawan Honggowib owo, Hero Wintolo, Yuliani Indrianingsih, Regina IMaula

Adiba

: a. N a m a Jurnal : IJASEIT (International Journal on Advanced Science, Engineering and Information Technology) b. N o m o r/V o lu m e : 2/10

c. Edisi (bulan/tahun) : 2020

d. Penerbit : IN S IG H T - Indonesian Society for Knowledge and Human Developmen

e. url dokumen :

http://insightsocietv.org/oiaseit/index.php/iiaseit/article/view/8407/pdf 1360

K o m p o n e n Y a n g Dinilai

Nilai M a k s i m a l J u r n a l I lm ia h Nilai A k h i r Y a n g D ip e ro le h I n t e r n a s i

o n a l B e r e p u t

asi

Q

I n t e r n a s io n al

□

N asi o nal T e r a k r e d it a

si

□

N a s i o n a 1 T i d a k T e r a k r e d ita si

□

Nasi o nal T e r i n

d e k s D O A J

□

J u r n a l N a s i o n a l T e r a k r e d i t a s i K e m r i s t e k d i k t i P e r i n g k a t

1

□

2

□

3

□

4

□

5

□

6

□

a. K elen g k ap an u n su r isi buku

(10%)

4 b. R u an g lingkup dan ked alam an

pem bahasan (30 %)

12

% 2-

c. K ecukupan dan k em utahiran d ata/inform asi dan m eto d o lo g i (30% )

12

3,z

d. K elengkapan un su r dan k ualitas penerbit (30% )

12

0j Z

T o ta l = ( 1 0 0 % ) 40

3o,S>

K o n t r i b u s i P e n g u s u l ( P e n u lis K e d u a d a r i T ig a P enulis )

(1 3 ,3 % x

^

... )

K o m e n t a r P e e r R e v ie w

1. T en tan g k elengkapan un su r isi b u k u ... ...

...,... ...

2. T en tan g ru an g Iingkup,dan k e d a la m a i pem bahasan 1 / i r v ^ U v f P £ V f ~ v t i L

3. T en tan g kecukunan daq kem utakhirun data/inform asi dan m etodologi.

... 7 ^ ^ l ( eh -W i* ' f t

4. T en tan g K elengkapan un su r dan k ualitas n en erb it y & > a r t{ $ £ y \ * u

cJ&rtan (M ^l <\ tfQ4lirb(t

,

Or.

e ^ v Uj« s.s,

Unit Kerja: u*. - W y