Summary of Recommendation System's Algorithms

On the lecture 7, Rosanna introduced the details of recommendation system to us. After class, I browsed the Internet and summarized the category of mainstream algorithms.

Firstly, the item-based collaborative filtering and user-based collaborative filtering, which have been taught in the lecture by Rosanna, are the core algorithms used by the large web sites, such as Amazon.com and JD.com. Shown as Figure 1, the process of collaborative filtering (CF) includes two main steps, the forecasting and recommending. The disadvantage of Item-based CF is the lack of diversity because the item are alike. However,     since the group of similar users is very sensitive, the drawback of the user-based CF is that it should calculate the matrix of similar users frequently so that the computation will be very big.

Figure 1 the process of collaborative filtering

Secondly, the content-based algorithm, which is basically depending on the text mining. For example, we can extract the keywords of two texts which you want to analyze and calculate the frequencies of the key wards. After that, you can calculating the texts’ similarity. The strength of this algorithm is no data is sparse.

The third algorithm I want to introduce is that K-Nearest Neighbor (KNN). K-NN is a type of instance-based learning and the K-NN algorithm is among the simplest of all machine learning algorithms. Both for classification and regression, the algorithm is used to calculate the contributions of the neighbors which means the nearer neighbors contribute more to the average than the more distant ones.

What’s more, the Slope One algorithm is another algorithm that I want to introduce. The slop one algorithm, in my opinion, is an easy method to fill the blanket places in the user-item rating metrics. For example, as the Figure 2, User X and Y have given marks to item1 and item2 while User A just rated to item1. So, what should the mark User A gives to item2? By using the Slope One, the mark should be: 4 - ((5-3) + (4-3))/2 = 2.5. This algorithm is very simple so the short coming is that the item the system recommends to you is common but not individual.

Figure2 example for the Slope One

Reference:

1. http://blog.csdn.net/huagong_adu/article/details/7362908

2. http://zh.wikipedia.org/wiki/Slope_one

3. http://blog.csdn.net/pi9nc/article/details/9068437

The problems which Social Network Analysis can solve

In the course of last week, Rosanna introduce the concept about Social Network Analysis (SNA) to us. In my opinion, the Social Network is an essential component in social media, which can be shown as a graph as below. The graph illustrate the links among people in groups.

  Social Network

 

The term 'relationship' is the bidirectional property while the mainstream social science concentrates on the single attributes, such as income, age and gender. There are several bidirectional attributes in the social activities, including bloodshed, social roles, emotional relations, cognition relations, action relations, flow relations, distance relations, similarity relations and co-occur relations. With these night bidirectional attributes, SNA can build a relation network and analyze the relationship on a pair of actors.

 

The SNA mainly consists of two sections—the visualization analysis and measurement on the relations of society, economics and emotion. If you want to use visualization analysis, you can choose the software as NetDraw, Pajek and Yed while you may use the Ucinet, NetMiner or Negopy to solve the measurement problems in SNA.

 

However, to be honest, which problem can be tackled by the SNA? In my opinion, it may help to solve the problems as follow:

1.       Interpersonal communication problems.

2.       Small world phenomenon.

3.       Association analysis in data mining.

4.       The meaning of the text output.

5.       Intelligence analysis on your competitor.

6.       The meanings of symbols.

7.       Analysis on correlation matrix or difference matrix.

8.       Knowledge management and knowledge transfer.

*Reference:

http://shenhaolaoshi.blog.sohu.com/136119138.html

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



 

 



Some thoughts of the Social Psychology

The lecture five, which is a self-learning lecture during the National Day Holiday, is about the Social Psychology. What is Social Psychology? It is a portion of psychology but study how human beings influence and communicate with each other particularly.  

In my opinion, I cannot agree more with the concept that human are born to be social. It means that human have the ability to communicate with the environment even when they are infants. For example, when a baby boy is hungry, he will express his feeling by crying in order to attract his parents’ attention. Another example is, a girl who was brought up by a wolf turned out to live like a wolf when she grew up in 1970s, which shows that human can interact with the environment even he is just a baby.

Furthermore, the way that human communicate with each other may be affected by the development of the technology. The lecture five refers to the topic about the SEC and indicates that the way people grasp knowledge has been changed by the online social network and the media. I think it is the truth because nowadays people can get the answers by the machine directly like the calculator and computer instead of gaining them by people. The three types of social Interactions in online world include human-to-human (OSNs), human-to-machine (cloud computing) and machine-to-machine (multimedia).  It is no doubt that the OSNs help humans grasp knowledge more efficiently. However, if the artificial intelligent appears in the future, the way for human beings to get knowledge may be more efficiently than nowadays.