We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
THE IMPACT OF DOMAIN-SPECIFIC STOP-WORD LISTS ON ECOMMERCE WEBSITE SEARCH PERFORMANCE.
- Authors
White, Barbara Jo; Fortier, Jenny; Clapper, Danial; Grabolosa, Pierre
- Abstract
Search time is an important determinant of e-service quality. Consumers may abandon slow websites prior to purchase. Much like an index in the back of a book facilitates searching, a similar index, called an inverted index, facilitates searching in a digital environment. Smaller indexes result in faster searches. One way to accelerate search time is to reduce the index size by removing common words like "the," "and," or "with." These words, called "stop words," offer little searchable meaning. Words on standard stop-word lists are frequently removed from indexes. However, e-commerce sites also contain high-frequency, low-value words that are domain-specific and should be added to standard stop-word lists. We created a corpus of over 36,000 eBay products in the furniture category and used linguistic analysis to assist in the creation of a domain-dependent stop-word list. We tested our domain-dependent stop-word list against a standard stop-word list and a control group. We experimentally generated furniture category queries and executed a random set of these queries against the three indexes and recorded the search times. A repeated-measures ANOVA was used to analyze results. As expected, the domain-dependent stop-word list resulted in the greatest reduction in search time. Implications for e-commerce sites are discussed.
- Subjects
EXCLUSION lists in automatic indexing; ELECTRONIC commerce; WEBSITES; INTERNET searching; INVERTED indexes; STOP words; ANALYSIS of variance
- Publication
Journal of Strategic E-commerce, 2007, Vol 5, Issue 1&2, p83
- ISSN
1554-5393
- Publication type
Article