Consumer markets are basically markets dominated by products and services designed for the general consumers. Consumer markets are typically split into four primary categories; consumer products, food and beverages products, retail products and transportation products. The consumer market pertains to buyers who purchase goods and services for consumption rather than resale. However, not all consumers are alike in their tastes, preferences and buying habits due to different in characteristics that can distinguish certain consumers from others. Industries in the consumer markets often have problems regarding deal with shifting brand loyalties and uncertainty about the future popularity sale of products and services.
Many of these markets on the Internet use collaborative filtering technology to assist and aid users. Collaborative filtering technique is a common method used to build personalized recommendations on the website. Some of the popular websites such as Amazon, iTunes and Netflix use collaborative filtering technology. In collaborative filtering, algorithms are used to make automatic predictions about a user's interest by compiling the user's buying behaviour. User's buying behaviour is the study of consumers and the processes they use to choose, use and dispose of products and services provided. In short, collaborative filtering is a technology used to recommend certain products and services to consumers based on their interest by referring the user's history of recent purchases on the website. The underlying assumption of the collaborative filtering approach is that if a person A has the same opinion as a person B on an issue, person A is more likely to have person B's opinion on a different issue X than to have the opinion on issue X of a person chosen randomly. For instance, a collaborative filtering recommendation system for television tastes could make predictions about which television show a user should like given a partial list of that user's tastes. Note that these predictions are specific to the consumer, but uses information from many other consumers. This differs from the simpler approach of giving an average score or rating for each product or service of interest, for example, based on its number of votes or its average rating of 1 to 5.
The growth of the Internet has made it much more difficult to effectively extract important information from all the available online information. The overwhelming quantity of data needs mechanisms for efficient information filtering. This is when collaborative filtering comes into play to solve the problem. The motivation for collaborative filtering comes from the idea that people often get the best recommendations from someone with similar tastes and interest to themselves. Collaborative filtering encompasses techniques for matching people with similar interests and making suggestions on this basis. Normally, the workflow of a collaborative filtering mechanism is a user states his or her preferences by rating the products in the system. These ratings can be assumed as an approximate representation of the user\'s interest and tastes in the corresponding domain. The system then matches the user's ratings against other users' ratings and finds the people with the most similar tastes and interest. With similar users, the system recommends items that the similar users have rated highly but not yet been rated by this user. One of the many popular websites that use this method is Netflix and iTunes. In Netflix and iTunes, consumers can set what movie or music genre the user would like to watch or listen and the system will find other users that match the same movie or music genre. The system will then suggest and recommend the movies or music that the other users like that the user has not watched or listened to.
Despite the benefits that collaborative filtering offers, there are also minor downfalls to the technique. Traditionally, data sparsity is seen as a key drawback of user-based collaboration filtering. A small number of co-rated items or no such ones between two users, resulting in unreliable or unavailable similarity information, and further incurring poor recommendation quality. It is often assumed that data sparsity may cause the lack of co-rated products. However, the analysis process is often not experimentally justified. The effects of the data sparsity on user-based collaboration filtering experiment with three steps to make a detailed interpretation. Firstly, investigating the relationships between the data sparsity and the number of the co-rated items. Secondly, exploring the characteristics of the number. Thirdly, evaluating the effects of the number on the recommendation quality. Experimental results show that the number of co-rated items does not drop, if data sparsity increases and as the number of co-rated items decreases, the recommendation quality does not drop. These results prove that the traditional analysis of the effects of data sparsity is problematic to the system. In short, lack of a number of rating of a product or even none leads to poor recommendation quality on the website because there is not enough information the website can compile to make high-quality suggestions.
The collaborative system can be used to help the students search for books in the college library. The interest and tastes of the user are the type of course or subjects the students pursue. The system will use the data given to search for books of their choice and it will give a recommendation based on that. For example, if a student wants to find a book on interior designing to do some research, the system could list down every book that has to do with interior designing.
In conclusion, collaborative filtering helps to suggest and recommend products that are new and unknown by the consumers based on the user's interest and tastes, for instance, favourite genres and favourite artist or actor. Collaborative filtering system helps to promote the new products on the website so it can maximize the potential sale of the new item. Furthermore, the system helps consumers to identify the quality of the item for sale by viewing its average rating which is rated by the consumers themselves. Although there are many advantages to the collaborative filtering system, there is also a minor drawback. The collaborative filtering system cannot produce good quality recommendations and suggestions if there are an only small amount of data or none that is available. The collaborative filtering system is very useful to the consumer marketing as it provides many benefits and maybe the system can be further improved in the near future.
...(download the rest of the essay above)