Friday, June 5, 2015

Gold in the Data

As Facebook rise, social network become the focus of attention once again. Compared with the traditional BBS, blog, social network is bridge to link a virtual world and real world. From the perspective of the classification of social network, Facebook, Twitter, LinkedIn represent three different kinds of social networks. Facebook social networking is based on the strong relationship between friends, help to friends of the relationship between maintaining and improvement; Twitter is based on one-way attention weak relationship of the social network, the network is conducive to shape the opinion leaders and the spread of news; LinkedIn is vocational social network for business people, help the user to use social relations for business communication and recruitment.

This three kinds of social networks have every day a lot of User Generated Content, and with unprecedented scale and mass, attracts numerous researchers to discover valuable information from the disorderly data. It's like frequently in probability and statistics for example, calculate the probability of the front side of the coin from throwing in the result it is difficult to see law of several times, but by tens of thousands of times of throwing experiments, it is easy to see that both the front and the occurrences of nearly equal. Social networks generated a lot of scale, social data, attracted including computer science, psychology, sociology, journalism and communication in the fields of experts and scholars to study and explore, hope to be able to use a stronger social network analysis and find out more processing power of human has not been explored.

For a wide range of social network analysis and research, there are many interesting research topic. Identification of a circle, for example, in a social network Community Detection, social network calculation, the influence of characters in the transmission of information on social networks model, false information and identification of bots, based on the social network information for the forecast of stock markets, elections, and infectious diseases, etc. Social network analysis and study is a cross field of discipline, so in the process of research, we usually use of sociology, psychology, and even basic conclusions and medical theory as a guide, through the use of machine learning in artificial intelligence, such as graph theory algorithm of social network behavior and simulate and predict the trend of the future.

For me, the most interesting thing is use SNA to predict the future. Social network to attract hundreds of millions of people every day on the Internet to release their own data, status, mood, these data gives us the opportunity to discover something we need to know. Such as by monitoring the Twitter public sentiment data, found the mood of the public data and many social phenomena and events have strong correlation. For example, some researchers found that regardless of whether they are "hope" positive emotions, or "fear" of the embodiment of the negative emotions are heralds a drop in the stock market index. Researchers believe that as long as there is in a social network public mood suddenly changed, will reflect on the uncertainty of the stock market, so the signal can be used to predict the future direction of the stock market.

However, our ability to predict for using social network data is also not too optimistic attitude, because of the social network prediction is based on the huge amounts of data, but for the massive text data has yet to reach the ideal accuracy analysis of the algorithm. Especially for judging from text messages for mood this seemingly simple question, its essence is the overlapping problem of unity and natural language processing and emotional psychology. But the current main methods of natural language processing by using the method of probability and statistics, as well as the lexical and syntactic analysis. The text emotional judgment is also based on thesaurus and grammatical structure of judgment and the method based on machine learning. However these methods for a bit complicated, especially with irony, and it is difficult to effectively determine the implied meaning of the language. In addition, for the use of social networking groups can not completely represent the effective mass, because people use social networks and age, regional, ethnic, and so on have very big difference, so only use social network on the data of prediction is likely to be biased and the final result, so the scientific and effective sampling method from the viewpoint of people for social network prediction is also particularly important one link.

(Posted on behalf of Bevis Zhang)

Wednesday, June 3, 2015

Using Social Network Analysis in performance evaluations

Given that the course’s name is “Social networks in organizations” I started thinking from the day we were assigned the task of creating a blog post, in a use of this methodology of understanding how groups of people are connected in, well, an organization. It wasn’t an easy thing to do I must admit. Even after several days into the course I was struggling with how something that prioritizes the “how” over the “what” can be used inside an organization.

But then, in a struck of inspiration, almost like the one I think hit Stephen Hawking into thinking and proving that a massive black hole is sitting in the center of our galaxy, I thought: Organizations live and breathe for the “How” over the “What”, and there is not one instance where this is more truth than the beloved mid-year and end of the year performance evaluations.

For you readers that haven’t been members of a medium to big sized company (anything above 20.000 employees qualifies as that), performance evaluations can be summarized into two things: Either arguments to promote you, or arguments to fire you, that’s the untold truth.
So where exactly does SNA fits into this picture? There’re a set of metrics in almost every performance score card, those coming from something called “360 review”. This is nothing more than how is your performance perceived by the people you interact with. Depending on the organization, the specific weight of this may vary.

 In the organization I was member of, it was hands down the most important metric, you could be the very best of your department, but if your 360 review was not in those same levels, you would then be a member of the infamous “high potential candidates” group. Think of them as the 99 MPH triple A prospect that no one in the bullpen likes.

How you are perceived and connected matters in an organization. Regardless what you talk about. The more connections you have, the more the number for the 360 review will be. Because the methodology of it rewards people well known by the managers performing the evaluation. And when more people are able to provide relevant information and comments about you, the more are the chances to obtain a good score in these metrics.

This is where SNA comes into play. Determining how really connected a person is before starting the assessment will give a little bit of validation to something that, in my opinion is very very wrong. Using cliques and degrees of connection as a way to determinate your true social network and establish whether or not it can create a positive impact in your performance and the way you are perceived. Using SNA could give the HR department a better view of how a person’s connections are consistent with the results of their personal evaluations before the 360 review. Let me make this clearer, because I believe this is the main benefit and the base argument of this post.

Suppose you have a high score in the personal review, you’re expected to have pretty much the same result in the 360. However this is not always the case, because unfortunately the reality is that if you’re working hard to get results, you can’t socialize and built your network, and people will perceive you as “high potential” but not connected, therefore, a bad candidate, soon to be “encourage to leave”. But what if after the SNA on this person the results are that under different attributes and networks, the person does have a strong connection with peers and it is perceived as a valuable contributor to the organization.

As I said, SNA can give some validation to an almost unfairly subjective process. It can provide evidence to determinate how any member of an organization is truly being a member, pretending to be a member or just not involved at all. Now, those pretending to be a member are the ones getting all the promotions, because like I said, is just almost impossible to be good at your job and building a strong network trough out the company. I believe that with SNA, the wide part of the curve will move towards the people that are contributing and making an effort into creating a beneficial network, for them and for the organization.


Tuesday, June 2, 2015

Using Social Network Analysis to analyse spreading pattern of backyard chickens in Bangkok’s boundaries in Thailand

1. To identify the social networks of the Avian Bird Flu in rural areas in Thailand.
2. To compare and examine the pattern in the spread of Backyard Chickens in different provinces in Thailand.

Situation and Problems
        Avian influenza cannot be diagnosed by symptoms alone, but it is diagnosed by collecting a swab during the first few days of illness. Human cannot  get avian influenza from properly handled and cooked poultry and eggs. The virus can infect a wide variety of avian species and mammals. It also affect to human health. The virus was officially reported the first time in Thailand in 2004. The number of patients was increasing at that time. The lack of biosecurity in chicken farming has showed as the factor and cause in the spread of virus in Thailand.
        SNA aims to identify and analyze the pattern of spreading of Avian Bird Flu in Thailand from activities related. Therefore, it could examine the pattern of diseases and lead to intervene the spread cycle of disease from related activities and potential people who could connect disease. It would help to identify the relationship of patterns and activities. 

Action Plan
Materials and Methods:
       I decide to do the questionnaire and clarify the questions before starting the data collection. According to the researches, I boundary the scope of the network to farmers who raise chicken both farmer’s household and farmer for merchandise. They have higher possibility to infect Avian Bird Flu than others. The questionnaire would contained both open and close questions to identify the movement and all related activities that lead to Backyard Chicken. Moreover, the research could also identify the risky location and ties all related activities to connect and disconnect the risk of the spread of virus. 
Data Collection:
     Open and close questionnaires collecting from the risky areas.  The scope of sample information:
1. Farmers and cock’s owner
2. Risky Area
3. Age

Result Analysis

        SNA will help to identify the key players, which can rapidly control outbreaks. The spread of disease can be limited by removing infected nodes which will break down the network. SNA will assist the authorities to implement more effective dis-ease control measures despite restricted manpower. 

Monday, June 1, 2015

Making decision on the best successor to family business enterprises by assessing the internal networks

Introduction and definition of Family Business
We as a business school student have dealt with a great deal of business cases that describe non-family businesses and gained deep insights of them by applying the framework of business disciplines. However, a large proportion of business in the world is actually established in the form of family business. It is reported by Family Firm Business Inc that family enterprises are amounted to 2/3 of all of the businesses all over the world, which have created over 50% of job opportunities in the vast of countries. 

For the sake of building a clear understanding of family business, I will define the concept of family business firms as following:

Family Firms are those in which multiple members of the same family are involved as major owners or managers, either contemporaneously or over time (Miller, Le-Breton Miller, Lester, Canella, "Are Family Firms Really Superior Performers,” Journal of Corporate Finance, Vol. 13, Issue 5, 2007)

The key to sustainably prosper family business is believed to be an effective handover process of management to the next generations leader. However, a number of the current family business owners do not have a clear guideline of how to identify the next leader who is much likely to contribute to the sustainable prosperity of the firm. This issue is complicated since the family business owners have to consider the relationship between each of the successor candidate and other employees as well as the successor's managerial prowess. Furthermore, this successor is supposed to be selected within a member of the owner's family. In order to match the candidates to the criteria above, I personally believe that SNA is an effective way to choose the appropriate successor since it visualize the internal organizational connections with each candidate's managerial ability. In this essay, I will take an example of family business and identify the problem within the firm, and then explain how SNA should address that problem. 

Hypothetical Situation
Let's suppose that there is a family business owner who runs a nursing homes in a super-aging country of Japan and currently ponder how to take over his ownership and managerial position to one of his family members. The brief description of the firm is as following:

• The nursing home is a facility where elderly citizens with senile dementia live together in support of licensed nurses.

• The firm is enjoying a financial success due to a rapid pace of aging society in Japan and its great dedication to building a good relationship with the adjacent communities.

• While the business outlook for the firm seems to be robust, it is faced by new competitors with low cost and needs to differentiate from them by enhancing the quality of its hospitality through training nurses.

• The firm comprises of 5 major departments: Finance division, Marketing devision, Sales division, Human Resource division, and nursing service devision.

• The owner has 3 candidates for the successor from his family, each of who has different business set skills and built different relationship with other employees in the firm.

Challenge Identification
The challenge here is to identify the best successor to the family firms who has an ability to enhance the quality of hospitality confirmed by the successor candidates' expertise and to gain a support from other employees easily by leveraging SNA. The SNA methodology will help the owner in the decision making process in the following ways:

• Get an understanding of the interpersonal connections of the firm.

• Provide the extent to which each employee has a reliance on the candidates by observing the degree connections

• Offer the owner information about which candidate has the required business skill sets to achieve the current business objective mentioned above.

• Identify who will help the new successor to be stabilized in the firm and how much influential power that supporter has.

Required Data
In order for the owner to select the best successor, I have to provide him with some data which are beneficial to increasing the accuracy of the selection process as following:

1. What is your age group? (Choose one group from: 20-30, 31-40,41-50 and over 50)
2. What is your gender?(Male or Female)
3. Are you a family member of the owner's or an outsider?
3. In which division are you currently working?
4. Which division is your previous division?
5. In which division do you have the longest working experience?
6. Who do you ask business advice for?
7. Who do you go for lunch with on a daily basis?

In my understanding,  the data above is relatively easy to gain since the questions just ask the respondents about their patterns of behaviors and therefore they can answer the given questions without having a sense of threat of promotion.

Crucial Measurement 
In order to help the owner select the best successor, I personally think that the most critical measurements are as followings:

• Eigenvector
-  This attribute measures the extent to which a person has connections with the others who have a plenty of connections. In order for the new successor to be able to communicate with employees to convey his direction, it is crucial for the successor to have a high eigenvector points so that his intention is carried to entire organization efficiently.

• Degrees
- This attributes measures how many nodes a person has in the firm. Given the intimate relations within the family business, the successor should have a number of degrees in order for the employees to familiarize with the new leader. The high degree points will help the successor to gain support in both personal and professional situations.

• Ego networks
- Ego network is an effective way to look at individual's relationship by focusing on a specific node. This helps the owner to grasp the extent to which each of the successor candidate has helpful supporters and credibility from others. More specifically, the ego networks clarifies whether the individual relationship is mutual or not.  Much attention to ego networks also provide the owner whether each of the successor candidate has more outsider-employees oriented or owner's family oriented. If the candidate shows a strong connection to either outsider-employees or owner's family oriented, he will have difficulty in harmonizing the firm. The ideal successor should be a person who has a balanced connections with these 2 groups.

• Cliques
- Cliques measures where a node(employee) belongs to a specific group. The measurement is beneficial as for identifying each of the successor candidates' ability to communicate with employees within a specific division. If the candidate have a closely tied relationship, that means he excels at communication skills confirmed by his business expertise.


With the help of SNA in decision-making process of the successor to the family business, the owner will be able to place the best successor who can exert significant influence on the critical employees or division to attain the current business objectives. In this case, the objective would be enhancing the quality of the hospitality by concentrating on the nursing training program. In order to attain that goal, the successor would be required to have a profound insight of the real situation of the nursing home, which is a solid knowledge foundation of nursing confirmed by working experience in that division. Based on this, I personally suggest to the owner that the successor should satisfy the following criteria:

• High degrees to the nursing service division to identify the problem in the hospitality and launch a effective implementation plan to enhance that. In order to ensure the successful business plan, the successor will be required to have a good relationship with the nursing service division, which is represented by the high number of degrees in the context of ego networks.

• The successor needs to have a high eigenvector so that the information flow of his idea gets lubricated. If the successor has a high eigenvector, the person will be able to gain valuable information from his subordinates and to give a direction to them in a timely manner.

• The successor should belong to the clique where the density is relatively high compared to the others.

Yuki Sakaguchi, MIB

• PWC, family business survey, "key findings", obtained from URL:

• Family Firm Institute, Inc. "Family Business Statistics from Around the world", obtained from URL:

Corrupt police network in India – Making the invisible to be visible


Applying social network analysis techniques to map out the police corruption in India, that will further prevent corruption and foster investigation. This analysis can be done using various data that is available from the commission of enquiry department,  which exposes the network and provides insights about the dark networks of police corruption.


Its not surprising today that, public officials who put-forward themselves before those of public have corrupted a system in which they ignore to act as public servant. Major form of corruption – Bribe (money in form of cash/kind/gift to carry out illegal action or dishonesty). Corruption channel in the indian police department flows right from the constables to the high rank officers. The most surprising truth is that, the police takes bribe from both, the accused and the complainants. The power that police enjoys is so wide that, they can harass, accuse, arrest even an honest individual. A weekly routine of a dishonest officer goes like – beating the poor people (e.g. auto-rikshaw pullers, street vendors, labourers etc.) by taking a percentage of their hard earned money, popularly termed as ‘hafta’ (weekly fixed amount).
The bribe rate of police department varies – for constables the rate is between Rs. 10 to Rs. 2000, for a sub-inspector it is Rs. 2000 to Rs. 10000 and for a deputy supritendant it is Rs. 10,000 to Rs. 20,000 or more.


It can be said that the corrupt police network has a similar culture and characteristics that match other types of criminal networks. I say this because, in the criminal network, the drug trafficking operations are carried out purely on kinship and friendship networks to help them protect from from law agencies. They deal only with people whom they trust and know. Therefore, it can be said that corrupt police network is also a trust based network/organisation. It is important to note that there are three possible key variable of bribe that are central to the police force – Bribes, transference of bribe and corrupt support


HYPOTHESIS 1:- The police network will extend to only those who can be trusted through corrupt association by a common subset. The people under this network perform specific roles who fill the spaces between nodes. Therefore, it is likely that a few key people within the network who facilitate communication.

HYPOTHESIS 2:-  The structure of a corrupt network will mirror the hierarchy of the police force and thus it will tend to have a high density.

HYPOTHESIS 3:-  The structure of the network would be highly centralised, which would reflect the beuracratic structure of the Indian police force. This would be visible in the long average path length as the communication in that network will follow the chain of command and hence it will have a high rating of centrality.

HYPOTHESIS 4:- SNA will be able to identify key influencers within the network that can be exposed and disrupt the entire corrupt network.

  • First, extract and gather police enquires and plot the corrupt networks as network maps.
  • Examine the network.
  • Use SNA to identify the strengths and weaknesses of the corrupt police network and develop plans and strategies for investigation and prevention.

  • Data sources collected from police enquiries could be imperfect
  • Network maps may mislead by plotting relationships that have been identified by a third party without any confirmation.

  • Density – This will help us show how close knit the entire network is and show the overall the strength of the network as well.
  • Centrality measure – It helps to give an indication of how concentrated a network is – Is the flow of resources controlled and managed by a small number of people or is it widely distributed? A high centralisation score indicates that some nodes in the network have more connections than others.
  • Average path distance – This measure helps to identify the level of ease of navigating around in the network. In other words, amongst people who are connected, how much do they have to travel to reach another person they are not connected to.
  • Reciprocity – Indicates whether there is a two-way or reciprocal relationship between the nodes in the network

Alok Agarwal
MBA candidate 2015
Hult International Buiness School