| 12.22.2008 21:06:56 |
 Jack Yang |
Comparison of Online Recommendation Functionality
Getting recommendation is more and more encountered in online experience. Browsing a site may quickly bring suggestions, like Amazon, eBay and Yahoo! Sometimes the recommendations are made by robots. Sometimes they are made by real people. Who makes the recommendation is not so important. But a successful recommendation must have two points.
1. Real experience with the recommended object. Recommender must know the object details and must have real experience of using the object.
2. Honesty and objectivity. Recommender shall not simply say that an object is “good” or “bad”. Instead, s/he shall point out “what the good (or bad) things are”. Recommender shall respect the truth, neither concealing the defects nor exaggerating the capability.
With more stores open online, techniques of online recommendation also grows quickly. Internet is good at speed and massive coverage. So it will be a good source for customers to refer to for making purchase decision. However, a site making recommendation must assure the above two points so as to survive.
Amazon, eBay and Yahoo! all give product recommendations. At Amazon, you will find what have been bought by readers who are interested in the same books you are browsing. On Yahoo! Auto category, for example, there is Editor’s Picks where a lot of new car reviews and auto shows are posted. It has Y! Groups designed to discuss for Hybrid cars, but obviously there are many spam messages inside. On eBay, there are Hot for Holiday product list and Great Finds and More Great Finds lists. eBay also have Community hub, where members can meet with one another via discussion boards, chat rooms, Online Workshops, eBay Groups, and the Answer Center for interactions.
Other sites not run by online stores, like acobay, is designed to promote sharing of real user experiences. These sites relates with product recommendation. Unlike online stores or networking sites, acobay is a consumer network that enables users of same products to connect and communicate with each other. Users can write review of products that s/he purchased. And they can also give ratings to the products. To make sure that the reviewers have real experience of purchase (rather than an advertising agent in disguise), Q&A function are added for the browsers. Questions can be asked for details (e.g. price or store address) that interests the browsers. Since these questions are random, only those who have real experiences can make the correct answers. As long as these questions can be answered satisfactorily, it will not matter too much even if the review was posted by an advertising agent because the agent has already known enough about the product.
The non-seller nature of these sites makes it unnecessary to solicit users to give good or bad comments. Independency is helpful to generate objectivity. And ranking is calculated on a massive review basis so as to avoid significant system bias.
free pictures
|
|
