Collaboration Analysis of Semarang City Dengue Hemorrhagic Fever Health Surveillance Officer with Social Network Analysis

This study aims to determine the pattern of a network collaboration between dengue hemorrhagic health surveillance personnel in the city of Semarang and to understand the flow of information deeply. The method of social network analysis / SNA (Social Network Analysis) on formal and informal communication networks between surveillance officers aims to produce sociometry and sociogram data so that the centrality of each actor in the network can be known. Interaction between Officers will be known through the analysis of the centrality of levels (degree), closeness (closeness), and intercession (betweenness). The approach used is descriptive quantitative. Data was collected using a research instrument in the form of a questionnaire, while the process of input and data analysis was carried out by looking at intact networks and ego networks carried out using UCINET. The results of the analysis show that officers dominate the centrality of the level and the intermediary with the position of Coordinator both City and District (Id. # 128, # 1, # 2, # 3). Collaboration based on working areas has a strong bond because 71% of the districts have a network density above 50%. While the value of closeness is dominated by surveillance members with id # 54, # 15, # 2, # 214, # 15 and # 10.


I. INTRODUCTION
Examples of the use of the SNA method in several studies, including research [2] in Batik Winda Sari UKM in Sragen City, concluded that individual brokers influence the relationship that occurs in these UKMs. In Research [3] [4] SNA shows that the relationship between students is not always based on academic issues, but also non-academic, the position of students in the network also determines success in examinations/studies. Research [5]. See the process of disseminating information about the "Wonderful Indonesia" country branding on the top platforms of the social networking sites Google Plus, Twitter and Facebook to become later input for the Ministry of Tourism of the Republic of Indonesia to increase the spread of country branding and the "Wonderful Indonesia" tourism campaign. Research [6] uses social network analysis to explore patterns created by the interactions of online users on Facebook during a disaster response. The study results show that social networks consist of three entities: individuals, emergency agencies, and organizations to share information in a state of disaster emergency. SNA research [7] was used by the European Commission to look at stakeholders who influence land-use decisions related to greening infrastructure at local and regional levels.
The SNA method can also be used to see collaborative research between lecturers in tertiary institutions. The resulting analysis can provide recommendations to improve the quality of research [8] as well as increase the Chancellor's expertise in recognizing the perspectives of others to anticipate emerging and systemic changes. Identify who is involved in discussions about teaching and learning and topics covered [9].
Social Network Analysis (SNA) is one of the most popular methods used to view social networks. SNA is used to analyze and map the interactions of individuals or institutions so that it  [12]. The SNA method is also used to analyze the network of sea container transport routes at the South Korean port of Incheon [13], analyze communication structures, interpersonal communication networks and the relationship of individual characteristics with interpersonal networks in social networks [14] [15] as well as distribution patterns and find key players in the distribution of pornographic content on Social Media [16].
The SNA method can also be used to study liaison organizations or actors who have the highest centrality and are well-positioned to bridge health and community services [17], collaboration between scientists, research centers, institutional network research, centers of excellence [ 18], e-learning studies [19], government programs in agriculture [20], viral events on social media [21], use of technology [22] and SME support institutions [23].
Based on previous research, this research was conducted because no research studies the collaboration of surveillance personnel / gasurkes in the city of Semarang. As for the studies on surveillance / gasurkes personnel, including: Research [24] which aims to determine the effectiveness of DHF surveillance officers in determining larval free rates (ABJ) in Semarang City. This study concludes that the DBD Gasurkes program in Semarang City has been effective by looking at the results of the value of effectiveness in each aspect compared to the percentage table of program effectiveness. Where the input aspect is effective (71%), the process aspect is very effective (92%), and the output aspect is effective (77%), although, in every aspect, there are still obstacles.
Research [25] aims to look for factors that influence the behavior of jumantik cadres in the early awareness system of DHF in Sendangmulyo Village, where someone who behaves well to the officers, has a significant influence and will realize good practices/behaviors in preventing DHF.
Research [26] focuses on analyzing the performance of Gasurkes in efforts to tackle DHF in endemic villages, where the results of the study showed that Gasurkes's knowledge was not good, there were good perceptions, lack of motivation, not yet enforced rewards, support of socioeconomic and political environment that was not optimal, the leadership process that has not been optimal, the workload that has not been evenly distributed, insufficient labor and sufficient facilities. The conclusion of the research shows that the performance of Gasurkes in the efforts to overcome DHF in endemic villages is not optimal.
From the results of the above study, no research has been found on collaboration between the gasurkes / DHF surveillance officers in Semarang City.

II. RESEARCH METHOD
Based on the description above, this study aims to analyze patterns of network collaboration or relationships between surveillance personnel / Gasurkes in Semarang City and to understand in depth the flow of information. SNA is used in this research to produce sociometry and sociogram data so that the centrality of each actor and its influence on the network can be known. A sociogram is a graph that illustrates the pattern of relationships and interests in social networks. A sociogram can describe the pattern of interaction between actors in a social network or the sociometric status of an actor in a social network or the overall state of actors in a social network [27].
Based on this, a mindset was formed, which later became the mechanism that would be used in the conduct of this research. The mechanism is illustrated in the following steps: : The process of data analysis uses the SNA method to measure centrality between officers, namely degree centrality, closeness, betweenness, and eigenvector. The attribute data is used to describe the collaborative relationship between surveillance personnel / gasurkes relational data in the form of symmetric metrics containing columns and rows containing the names of officers. Researchers put a relation number 1 if there is a link.
Otherwise, the number 0 is given if there is no link between the name in line and the name in the column. The data illustrations in Table 4 in the symmetrical form are as follows;

A. Network Density
Network structure can be seen through the calculation of network density. Network density is the comparison between the number of links (ties) with the number of links that might appear. Table 5 data shows that the overall network density (density) of the Semarang City Surveillance Network is 0.053 or 5.3%.  The results of UCINET calculations. Table 6 illustrates the overall network statistics. The total number of relationships formed in the entire network is 2485, with 215 employees. A maximum value of 76 indicates the number of actors who contacted or were contacted by each actor. Density-based in the work area can be seen in Figure 2.

A. Degree Centrality
Based on Table 7 shows ten actors with the highest degree of centrality value, where officer # 128 is the most popular central actor/actor in the surveillance network / gasurkes with 78 links, followed by actor # 1 with 65 links, actor # 2 with 63 links, actors # 3 with 53 links, actor # 55 with 43 links and followed by other actors. The implication is that the above actors have stronger capabilities in connecting other officers in the network. The sociogram of the degree of centrality in the surveillance / gasurkes power network in Semarang city in Figure 5 where

B. Closeness
The next data that can be taken is closeness, where the lower the value of closeness, the better it will be because it shows the low distance each actor has to deal with other actors. The shortest distance into (in fairness) is the shortest distance reached by the officer in sending information to other officers in the network. Table 9 shows that the officer who has the shortest distance in sending information to the surveillance network is officers number 54, 15, 2, 3, 167,  In Closeness value is interpreted as an indicator to measure the influence of an officer in the network, where the measure is a measure of how far information can be spread from one officer to another officer. Table 9 shows that officers who easily accept the distribution of information     The sociogram of betweness centrality is shown in Figure 7, where the size of symbols and labels is based on the betweness value of each actor/officer. In the picture, the actor in charge as coordinator (128, 1,3,2, and 55) has a greater symbol and label than the other actors. It illustrates the function of the coordinator plays an important role as an intermediary/information bridge.

Figure 7. SOCIOGRAM BETWENESS CENTRALITY OF SURVEYLANS OFFICER OF SEMARANG
The collaboration of officers based on the Work Area The success of the collaboration between officers in a sub-district area depends on the density of the network, where the denser a network/group, it will facilitate the information or knowledge spread within the group, it can be concluded the more relations are formed, the interaction between officers runs smoothly.  Secondly, seen from the analysis of collaboration between sub-district work areas, several sub-district areas tend to have components that have weak ties. Namely South Semarang, Gayamsari, Mijen, and North Semarang. Where it is an evaluation material for the Coordinator in each of these districts, because the success of the work area is very dependent on the density of communication in the region because the denser the area, the easier information or knowledge is spread within the region. If the area density is high so there is a large amount of communication between officers. The third, looking at the results of surveillance based on educational background, it is necessary to evaluate the recruitment of surveillance personnel because there is an imbalance in the number of personnels with an environmental health education background.