1. In a computing environment, a method comprising:
using a computer, determining a similarity score for a page based on similarity between a keyword for the page and page data of the page, wherein determining the similarity score comprises obtaining a first vector from the keyword, obtaining a second vector from the page data, and determining the similarity score based on a relationship between the first and second vectors, wherein the page data includes a page portion and a text portion corresponding to a first set of text, and further comprising, parsing at least part of the page portion into a second set of text, and wherein obtaining the second vector comprises processing the first and second sets of text based on term weighting;
determining a proximity score for the page based on proximity between the keyword for the page and the page data based on word order;
combining the similarity score and the proximity score into a relevance score; and
associating the relevance score with the keyword and the page.
2. The method of claim 1 wherein the page comprises an advertisement and wherein the keyword comprises a bidding keyword.
3. The method of claim 1 further comprising, receiving a query and using the relevance score to select the page from among a plurality of pages.
4. The method of claim 1 further comprising, determining whether the page portion corresponds to a content page, and if so, weighing the second set of text differently from when the second set of text does not correspond to a content page.
5. The method of claim 4 wherein determining whether the second set of text corresponds to a content page comprises computing a link density based on the page portion.
6. The method of claim 1 wherein obtaining the first vector from the keyword includes weighing at least one term of the keyword based on part of speech term weighing.
7. The method of claim 1 wherein obtaining the first vector from the keyword includes performing stopword removal on the keyword or performing stemming on the keyword, or performing both stopword removal and stemming on the keyword.
8. The method of claim 1 wherein determining the proximity score comprises obtaining a first word set from the keyword, obtaining a second word set from the page data, and using commonality between the first and second word sets to determine the proximity score.
9. The method of claim 1 wherein determining the proximity score comprises obtaining a first bigram set from the keyword, obtaining a second bigram set from the page data, and obtaining a value based on bigrams common to both bigram sets relative to a total number of bigrams in both bigram sets.
10. The method of claim 9 wherein obtaining the first bigram set from the keyword includes performing stopword removal on the keyword or performing stemming on the keyword, or performing both stopword removal and stemming on the keyword.
11. The method of claim 1 further comprising:
associating relevance scores for a plurality of pages;
receiving a query;
using the relevance scores to rank the plurality of pages; and
re-ranking the plurality of pages based on a proximity relationship between the query and the keyword, and, the relevance score of each of the plurality of pages.
12. In a computing environment, a system comprising:
a computer comprising a relevance score calculation mechanism that determines a similarity score for an advertisement based on similarity between a bidding keyword for the advertisement and advertisement data of the advertisement, wherein the similarity score is determined by obtaining a first vector from the bidding keyword, obtaining a second vector from the advertisement data, and determining the similarity score based on a relationship between the first and second vectors, wherein the advertisement data includes an advertisement portion and a text portion corresponding to a first set of text, and further comprising, parsing at least part of the advertisement portion into a second set of text, and wherein obtaining the second vector comprises processing the first and second sets of text based on term weighting, the relevance score calculation mechanism further determines a proximity score for the advertisement based on proximity between the bidding keyword for the advertisement and the advertisement data based on word order, the relevance score calculation mechanism further combining the similarity score and the proximity score into a relevance score and associating the relevance score with the bidding keyword and the advertisement;
a data store coupled to the relevance score calculation mechanism that stores the relevance score in association with the bidding keyword of that advertisement, and maintains other relevance scores for other bidding keywords of other advertisements; and
a delivery engine coupled to the data store to return at least one advertisement selected from the data store based on a query keyword obtained from the query and the stored relevance information associated with the bidding keyword of that advertisement.
13. The system of claim 12 further comprising a relevance verification mechanism coupled to the delivery engine that selects a set of advertisements based on relationships between the query keyword and the bidding keyword of each advertisement, computes a final score for each advertisement of the set based on the relationships between the query keyword and the bidding keyword and the relevance score associated with that advertisement, and ranks the advertisements relative to one another based on each advertisement’s final score.
14. One or more computer storage media having computer-executable instructions, which when executed perform steps, comprising:
determining a bidding keyword vector representing the bidding keyword for an advertisement and a page data vector representing the page data for the advertisement;
computing a similarity score for the advertisement based on a vector relationship between the bidding keyword vector and the page data vector;
computing a proximity score for the advertisement based at least in part on bigrams in common between the keyword and the page data;
combining the similarity score and the proximity score into a relevance score;
associating the relevance score with the advertisement;
associating relevance scores for a plurality of advertisements;
receiving a query;
using the relevance scores to rank the plurality of advertisements; and
re-ranking the plurality of advertisements based at least in part on a relationship between the query and the bidding keyword and the relevance score of each of the plurality of advertisements.
15. The one or more computer storage media of claim 14 having further computer-executable instructions comprising, determining a proximity relationship between bigrams of the query in common with bigrams of the bidding keyword.
16. The one or more computer storage media of claim 14 wherein determining the page data vector comprises weighing a term in the page data relative to other terms based on page content analysis, HTML attribute analysis, or part of speech analysis, or any combination of page content analysis, HTML attribute analysis, or part of speech analysis.
The claims below are in addition to those above.
All refrences to claim(s) which appear below refer to the numbering after this setence.
We claim:
1. A composition for inhibiting body odor, which comprises trehalose andor maltitol as an effective ingredient and a carrier.
2. The composition of claim 1, which further contains water andor alcohol.
3. The composition of claim 1, which further contains one or more members selected from the group consisting of maltooligosaccharides, cyclic saccharides, neutral polysaccharides, humectants, antibacterial agents, bacteriostats, germicides, antiseptics, antioxidants, ultraviolet shielding agents, ultraviolet absorbents, antiperspirants, vitamins, plants extracts, photosensitizers, emulsifiers, surfactants, feeling improvers, thickening agents, powdered materials, and refreshing agents.
4. The composition of claim 1, which is capable of inhibiting the formation of one or more ingredients selected from the group consisting of 2-nonenal, 2-octenal, and 2-hexenal as ingredients of body odor.
5. The composition of claim 1, which is in the form of a liquid, solid, semisolid, or gel.
6. An article which can be contacted with a living body and which contains the composition of claim 1.
7. The article of claim 6, which is a member selected from the group consisting of woven fabrics, non-woven fabrics, sponges, macroporous synthetic resins, cottons, and absorbent cottons.
8. The article of claim 6, which is a member selected from the group consisting of daily goods, landry goods, clothes, bedclothes, personal items, goods for hygiene, medical supplies, goods for carecaringnursing care, sport goods, cleaning goods, compositions for external application, compositions for scalp and hair, oral compositions, and goods for pets.
9. A method for inhibiting body odor, which comprises a step of contacting said composition of claim 1 with the secretion from a living body.
10. The method of claim 9, which comprises the steps of contacting said composition with an article which can be contacted with said living body, and contacting said composition with the secretion from said living body.
11. The method of claim 10, wherein said article is a member selected form the group consisting of woven fabrics, non-woven fabrics, sponges, macroporous synthetic resins, cottons, and absorbent cottons.
12. The method of claim 10, wherein said article is a member selected from te group consisting of daily goods, landry goods, cloths, bedclothes, personal items, goods for hygiene, goods for medical use, goods for carecaringnursing care, sport goods, goods for cleaning, compositions for external application, compositions for scalp and hair, oral compositions, or goods for pets.
13. The method of claim 9, which comprises the steps of incorporating said composition into water which can be contacted with said living body, and contacting said composition with the secretion from said living body.
14. The method of claim 13, wherein said water is hot water for bath.
15. The method of claim 9, wherein the secretion is sweat or sebum.
16. The method of claim 9, wherein the secretion exists on the surface of said living body.