Adaptive Facet Ordering for Online Shopping
Adaptive Facet Ordering for Online Shopping
ABSTRACT:
Faceted browsing is widely used in Web shops and product comparison sites. In these cases, a fixed ordered list of facets is often employed. This approach suffers from two main issues. First, one needs to invest a significant amount of time to devise an effective list. Second, with a fixed list of facets it can happen that a facet becomes useless if all products that match the query are associated to that particular facet. In this work, we present a framework for dynamic facet ordering in e-commerce. Based on measures for specificity and dispersion of facet values, the fully automated algorithm ranks those properties and facets on top that lead to a quick drill-down for any possible target product. In contrast to existing solutions, the framework addresses e-commerce specific aspects, such as the possibility of multiple clicks, the grouping of facets by their corresponding properties, and the abundance of numeric facets. In a large-scale simulation and user study, our approach was, in general, favorably compared to a facet list created by domain experts, a greedy approach as baseline, and a state-of-the-art entropy-based solution.
PROJECT OUTPUT VIDEO:
EXISTING SYSTEM:
-
The faceted search system proposed in existing focuses on both textual and structured content. Given a keyword query, the proposed system aims to find the interesting attributes, which is based on how surprising the aggregated value is, given the expectation. The main contribution of this work is the navigational expectation, which is, according to the authors, a novel interestingness measure achieved through judicious application of p-values.
-
These solutions often assume that there is a ranking of the results, based on a preceding keyword-based query or external data, which is often not the case for e-commerce
DISADVANTAGES OF EXISTING SYSTEM:
-
Large number of facets are available. Displaying all facets may be a solution when a small number of facets is involved, but it can overwhelm the user for larger sets of facets
-
Currently, most commercial applications that use faceted search have a manual, ‘expert-based’ selection procedure for facets or a relatively static facet list. However, selecting and ordering facets manually requires a significant amount of manual effort.
-
Furthermore, faceted search allows for interactive query refinement, in which the importance of specific facets and properties may change during the search session. Therefore, it is likely that a predefined list of facets might not be optimal in terms of the number of clicks needed to find the desired product.
-
This method is likely not to be suitable for the domain of e-commerce, where also small data sets occur and statistically deriving interesting attributes is not possible.
-
Approach does not consider numeric facets and the use of disjunctive semantics for values.
PROPOSED SYSTEM:
-
We propose an approach for dynamic facet ordering in the e-commerce domain. The focus of our approach is to handle domains with sufficient amount of complexity in terms of product attributes and values. Consumer electronics (in this work ‘mobile phones’) is one good example of such a domain. As part of our solution, we devise an algorithm that ranks properties by their importance and also sorts the values within each property.
-
For property ordering, we identify specific properties whose facets match many products (i.e., with a high impurity). The proposed approach is based on a facet impurity measure, regarding qualitative facets in a similar way as classes, and on a measure of dispersion for numeric facets. The property values are ordered descending on the number of corresponding products. Furthermore, a weighting scheme is introduced in order to favor facets that match many products over the ones that match only a few products, taking into account the importance of facets.
-
Our solution aims to learn the user interests based on the user interaction with the search engine.
ADVANTAGES OF PROPOSED SYSTEM:
-
In our study, we use the common disjunctive semantics for values and conjunctive semantics for properties and take into account the possibility of drill-ups. This means that result set sizes are expected to both increase and decrease during the search session, either by deselecting a facet or choosing an addition facet in a property
-
In terms of the number of clicks, our approach seems to outperform the other methods, except in the case of the Best Facet Drill-Down Model, where each approach performs equally well. Furthermore, for the Combined Drill-Down Model, our approach results in the lowest number of roll-ups and the highest percentage of successful sessions.
-
The relatively low computational time makes it suitable for use in real-world Web shops, making our findings also relevant to industry. These results are also confirmed by a user-based evaluation study that we additionally performed.
MODULES:
-
Facet Optimization
-
Search Sessions
-
Computing Property Scores
-
Computing Facet Scores
MODULE DESCRIPTIONS:
Facet Optimization:
In this Module, We distinguish two main facet types: qualitative facets (e.g., WiFi: true) and numeric facets (e.g., Lowest price (Rs):64.00). We further distinguish between two types of qualitative facets: nominal facets and Boolean facets. Nominal facets are, for example, those for the property Display Type, and can have any nominal value. Boolean facets are for instance Multitouch, and have only three options from an interface perspective: true, false, or No preference.
Search Sessions:
A query in a search session is defined as a collection of previously selected facets. We have decided to apply disjunctive semantics to a selection of facets within a property. For facets across different properties, we use a conjunctive semantics. For example, selecting the facets Brand:Samsung, Brand:Apple, and Color:Black results in (Brand:Samsung OR Brand:Apple) AND Color:Black. Several ecommerce stores on the Web (e.g., Amazon.com and BestBuy.com) use the same principle, which, from a user experience point-of-view, is very intuitive. Our approach assumes that users can undertake two types of actions: drill-down and roll-up. A drilldown is defined as an action of selecting one or more facets, leading to a reduction of the result set size. A roll-up action increases the result set size, which is likely to happen when the user notices that the selected facets are too strict.
Computing Property Scores:
The outcome of the property scores is used to first sort the properties, after which the facet scores, discussed in the next section, are used to sort the values within each property. The score for each property is computed separately and can thus be done in parallel.
Computing Facet Scores:
Our approach also sorts the values within each property in order to reduce the value scanning effort. For numeric properties, value ordering is neglected, as these are often represented with a slider widget in user interfaces. The slider widgets give an indication of the minimum and maximum values for a property, and allow the user to freely define a range of facets within these boundaries. For qualitative properties our approach employs the facet count, ranking facets descending on count, per property. As the target product is unknown to the system, this will increase the chance that a facet matching the target product is placed on top.
SYSTEM REQUIREMENTS:
HARDWARE REQUIREMENTS:
- System : Pentium i3 Processor
- Hard Disk : 500 GB..
- Monitor : 15’’ LED
- Input Devices : Keyboard, Mouse
- Ram : 4 GB.
SOFTWARE REQUIREMENTS:
- Operating system : Windows 10/11.
- Coding Language : C#.net.
- Frontend : ASP.Net, HTML, CSS, JavaScript.
- IDE Tool : VISUAL STUDIO.
- Database : SQL SERVER 2005.