1. Select ( Select operates on a TSR to produce a new one, selecting instances of resource items and OR components whose paths satisfy some defined predicates. Figure 3(a) shows an example involving a select operator in the query construct a TSR to include all cameras from TSR ‘SLR systems’, other than Pentax, with price greater than 200, having ‘/photo/35mm systems’ in their paths:
Abstract. Hierarchical structures are a way to organize and enrich semantically the available information on the Web. Popular examples of such structures are the product catalogs of e-market stores, which provide data (i.e. products) organized in thematic
<brand‘Pentax’,price>200><‘/photo/35mm systems’ $_>(SLR systems)
Results are presented in Figure 3(b). Notice that /photo/35mm systems $_ holds if there is a path p in an OR component, such that /photo/35mm systems p (i.e. the sequence of nodes in /photo/35mm systems appears identically in p). In this case, the whole OR component is retrieved.
Figure 3: select operator.
2. Project ( Project operates on a TSR to produce a new one, keeping only some of the paths of each OR component or OR components on the whole, and some of the attributes of the related resource item. Figure 4(a) presents an example of a project operator in the query construct a new TSR from the TSR ‘SLR systems’ keeping only the rightmost component (i.e. #2) and the attributes ‘model’ and ‘lens_id’ of the resource item:
<model, lens_id><#2>(SLR systems)
Results are presented in Figure 4(b).
3. Cartesian product (x). Cartesian product operates on two TSRs to produce a new TSR, combining (a) every OR component of the first TSR with every OR component of the second TSR, constructing an AND group of paths, and (b) every instance of the resource item in the first TSR with every instance of the resource item in the second TSR. Figure 5(c) shows the cartesian product of the two TSRs of Figure 5(a) and (b). The AND group of paths /photo/35mmSLR/bodies and /photo/lenses (i.e. the leftmost OR component of TSR SLR systems) is combined with the one and only OR component /camera & lenses/lenses of TSR Lenses, to construct a new AND group of three paths. Similarly, the rightmost OR component of TSR SLR systems is combined with /camera & lenses/lenses of TSR Lenses, to construct a new AND group of two paths. The new TSR has a resource item with attributes from both TSRs and combinations of instances.
Abstract. Hierarchical structures are a way to organize and enrich semantically the available information on the Web. Popular examples of such structures are the product catalogs of e-market stores, which provide data (i.e. products) organized in thematic
Figure 4: project operator.
4. Union (U) / Intersection (). Union operates on two TSRs with the same set of resource item attributes. It produces a new TSR with all the OR components of the two TSRs and the union of their instances. Similarly to the union operator, intersection operates on two TSRs with the same set of resource item attributes. It produces a new TSR with all the OR components of the two TSRs and their common instances. Figure 6(c) shows the union of the two TSRs of Figure 6(a) and (b). The TSR constructed has all OR components of both TSRs. Also, the result contains instances either from the first or the second TSR.
Figure 5: cartesian product operator.
5. Difference (-). Difference operates on two TSRs with the same set of resource item attributes. It produces a new TSR with the OR components of the first TSR and the instances of the first which do not exist in the second.
Abstract. Hierarchical structures are a way to organize and enrich semantically the available information on the Web. Popular examples of such structures are the product catalogs of e-market stores, which provide data (i.e. products) organized in thematic
Figure 6: union operator.
4. The PatMan System
We have developed PatMan, a prototype system that manipulates paths and raw data in hierarchical catalogs in a uniform way. PatMan provides modules to import TSRs from hierarchical catalogs coming in various forms (e.g. RDFs representation), and store them in a relational database system. It also helps the user posing queries with a visual interface that offers pictorial query-by-example capabilities. Finally, it can visualize the query results, showing the structure of the resulting TSRs as well as the raw data that satisfy the query predicates. Figure 7 shows the architecture of the system. In the next subsections we describe each one of system modules in detail.
Figure 7: System architecture.
4.1 Storage Module: Import and Store TSRs
Using the storage module, one can import TSRs from hierarchical catalogs coming in various forms, for example in RDFs representation. The module detects IS_A paths in RDFs schemas (i.e. class/subclass relationships) and constructs OR components for TSRs. Resource items are initially empty, but the user