Shopping advisor and other uses of knowledge base about products
In the post Yet another idea of collaboratively edited knowledge base I proposed to build a knowledge base on various topics. In this post I am going to present some possible applications of this knowledge base constrained to knowledge about products available on the market. Possible applications of the general knowledge base are described in the post General applications of the knowledge base.
Imagine an application that uses a knowledge base of products sold in stores to help consumers choose the right products (see the picture).
The knowledge base is marked in green. In this application we are only interested in a part of the knowledge base about products available on the market – initially only groceries. Information about products will be added to the database by producers, stores and consumers.
A consumer sets (in the application’s profile) criteria of products that are important to him, specifying their weights. He can choose from 4 groups of criteria:
- related to him, for example taste of food or whether food is healthy (e.g. is it a part of a healthy diet?, or does it contain harmful additives? etc.)
- related to the environment, for example: is production or packaging harmful to the environment?
- related to animals, for example breeding conditions
- related to other people, for example: is the product made by local manufacturers?, does the manufacturer pay taxes in a chosen country?, etc.
Then the consumer can go to a store and use his smartphone to scan bar codes of selected groceries. A list of scanned products will appear on his smartphone. Products that meet the selected criteria will be marked in green, products that meet the criteria but are much cheaper in another store will be marked in yellow, and products that do not meet the criteria will be marked in red. Additionally, if a product does not meet the criteria, the application may suggest a similar but better product.
The rules translating information set in a customer profile into assessment of product can be written by experts as plug-ins for the application.
We can also imagine another function. Stores collect information about products you buy using loyalty cards. Thanks to the General Data Protection Regulation you can ask stores to share this data with you via the Internet. Having this data, you can order for example monthly reports containing assessment of products you bought, listing the worst products often bought by you, and proposing better alternatives.
Similar databases and applications
There are already databases containing partial information about groceries, for example about their composition (amount of sugars, fat, proteins and additives such as preservatives and food colorings). However, most often they are not free – free databases are available for products sold in the United States, France and Switzerland – that is why using them in new computer applications can be troublesome.
In some countries there are applications that evaluate groceries taking into account only one criterion, for example food additives or manufacturer country. Unfortunately, there is no application that takes into account all criteria that are relevant to a consumer .
To sum up, we lack a holistic solution to the problem of choosing products. I think that this is due to a lack of a generally available knowledge base – that is why I proposed a method of its building in the post Yet another idea of collaboratively edited knowledge base.
List of similar databases, applications and concepts
- Open Icecat – a free database of products: electronics and some other types of products
- USDA Food Composition Databases – a free database of groceries available in the US
- Open Food Facts – a free database of groceries available in France, the US and Switzerland
- Brandbank – a paid database of groceries available in the US, the UK, Ireland, the Netherlands, Belgium, Denmark, Poland, the Czech Republic, Slovakia, Hungary and Thailand, made by Nielsen Brandbank, used by stores
- FoodSwitch – an application evaluating groceries composition; available in Australia, New Zealand, the UK, the US, South Africa, India and China; does not contain information about food additives 
- GoodGuide – an application evaluating groceries and cosmetics, available in the US
- multiple-criteria decision analysis
- Nutri-score – a method of evaluating groceries used in France, also implemented in Open Food Facts database
Other possible uses of the knowledge base about products
I could add to the system a list of products owned by me, for example: Peugeot 307 car, Bosch GSB 1300 drill, Toshiba 55V5863DG TV set etc. Thanks to this a manufacturer can easily inform me if, for example, he wants to call car owners to a workshop due to a manufacturing defect. Finding a spare part will also be easier because the knowledge base will contain information about spare parts compatible with a product I own.
Additionally, if I want to sell a used product, it is enough to select it from the list of my products, check an I want to sell option and optionally provide differences between my product and the new product resulting for example from its usage. Information about available used product will appear automatically next to new products after clicking the Buy button on the product overview page (see figure 4 in the post Yet another idea of collaboratively edited knowledge base).
Moreover, instead of choosing the I want to sell option, I could choose an I want to lend option. If someone needs a drill for a while, then maybe borrowing it from somebody living in the same estate will be more convenient. If there are no people willing to lend a drill at the moment, then the person willing to borrow it can subscribe to the drill term on the estate and he will be notified when the owner of a drill selects the I want to lend option. On the other hand, the drill owners living in the estate can be notified that someone has subscribed to this term.
 C. Chung, R. Proskuryakov, D. Sundaram: Sustainable Social Shopping System, proceedings of 6th International Conference on Computational Collective Intelligence, 2014
 E. Dunford, H. Trevena, C. Goodsell, KH Ng, J. Webster, A. Millis, S. Goldstein, O. Hugueniot, B. Neal: FoodSwitch: A Mobile Phone App to Enable Consumers to Make Healthier Food Choices and Crowdsourcing of National Food Composition Data