1. A process for separating xylenes starting from a feed comprising cuts of isomers of aromatic hydrocarbons containing 8 carbon atoms, the process comprising selectively adsorbing a xylene isomer from the feed in a simulated moving bed in the presence of a desorbent, using particles of agglomerated zeolitic adsorbent based on zeolite crystals with a number-average diameter less than or equal to 1.2 \u03bcm, wherein the number-average diameter of said particles of adsorbent is between 150 \u03bcm and 500 \u03bcm and the mechanical strength measured by the Shell method series SMS1471-74 adapted for agglomerates with a size below 500 \u03bcm is greater than or equal to 2 MPa.
2. The process for separating xylenes according to claim 1, wherein the number-average diameter of said particles of agglomerated zeolitic adsorbent is between 200 \u03bcm and 400 \u03bcm.
3. The process for separating xylenes according to claim 1, wherein the granulometric distribution of said particles of adsorbent is such that there is no particle with size less than 100 \u03bcm.
4. The process for separating xylenes according to claim 1, wherein the number-average diameter of the zeolite crystals is between 0.1 \u03bcm and 1.2 \u03bcm.
5. The process for separating xylenes according to claim 4, wherein the number-average diameter of the zeolite crystals is between 0.5 \u03bcm and 0.8 \u03bcm.
6. The process for separating xylenes according to claim 1, wherein the process is carried out in a simulated moving-bed unit having the following characteristics:
number of beds between 4 and 24
number of zones: at least 4.
7. The process for separating xylenes according to claim 6, wherein the cycle time, corresponding to the time between two injections of desorbent on a given bed, is between 4 and 18 min.
8. The process for separating xylenes according to claim 1, wherein the adsorption is carried out at a temperature from 100\xb0 C. to 250\xb0 C., and at a pressure between the bubble pressure of the xylenes at the process temperature and 3 MPa.
9. The process for separating xylenes according to claim 1, wherein the ratio of the flow rates of desorbent to feed is between 0.7 and 2.5 and the recycling rate is between 2.0 and 12.
10. The process for separating xylenes according to claim 1 wherein the xylene which is selectively adsorbed is para-xylene and wherein the agglomerated zeolitic adsorbent is based on zeolite X or LSX having an SiAl atomic ratio such that 1.0\u2266SiAl<1.5.
11. The process for separating xylenes according to claim 10, wherein the agglomerated zeolitic adsorbent further comprises:
i. a content of barium oxide BaO and a content of potassium oxide K2O such that the ratio of the number of moles of the total barium oxide+potassium oxide (BaO+K2O) to the number of moles of the total (BaO+K2O+Na2O) is greater than 90%;
ii. a content of potassium oxide K2O such that the ratio of the number of moles of potassium oxide K2O to the number of moles of barium oxide BaO is less than 0.5; and
iii. a total content of oxides of alkali-metal or alkaline-earth ions other than barium and potassium preferably below 5% relative to the total weight of the agglomerated zeolitic adsorbent.
12. The process for separating xylenes according to claim 10, wherein the agglomerated zeolitic adsorbent has a grain density between 1.1 and 1.4 gmL, as measured by mercury intrusion (expressed relative to the dry mass of the zeolitic adsorbent) and a total pore volume measured by mercury intrusion (pore volume contained in the macropores and the mesopores with apparent diameter greater than 4 nm) between 0.20 and 0.35 mLg (expressed relative to the dry mass of the zeolitic adsorbent).
13. The process for separating xylenes according to claim 10, wherein the process carried out at a temperature from 165\xb0 C. to 185\xb0 C. and wherein the water content of the hydrocarbon effluents is adjusted between 20 ppm and 150 ppm, by adding water to the feed comprising the cuts of isomers of aromatic hydrocarbons containing 8 carbon atoms andor to the desorbent.
14. The process for separating xylenes according to claim 10, wherein the desorbent is selected from the group consisting of toluene and para-diethylbenzene.
15. The process for separating xylenes according to claim 1, wherein meta-xylene is selectively adsorbed and wherein the agglomerated zeolitic adsorbent is based on zeolite Y having an SiAl atomic ratio such that 1.5<SiAl<6.
16. The process for separating xylenes according to claim 15, wherein the agglomerated zeolitic adsorbent further comprises:
i. a content of sodium oxide Na2O and a content of lithium oxide Li2O such that the ratio of the number of moles of sodium oxide to the number of moles of the total sodium oxide+lithium oxide (Na2O+Li2O) is greater than 65%; and
ii. a total content of oxides of alkali-metal or alkaline-earth ions other than sodium and lithium below 5% relative to the total weight of the zeolitic adsorbent.
17. The process for separating xylenes according to claim 15, wherein the process is carried out at a temperature from 120\xb0 C. to 180\xb0 C. and wherein the water content in the hydrocarbon effluents is adjusted between 0 ppm and 80 ppm, by adding water to the feed comprising the cuts of isomers of aromatic hydrocarbons containing 8 carbon atoms andor to the desorbent.
18. The process for separating xylenes according to claim 15, wherein the desorbent is selected from toluene and indane.
19. (canceled)
20. (canceled)
21. The process according to claim 10, wherein para-xylene is separated in a purity greater than or equal to 90%.
22. The process according to claim 15, wherein meta-xylene is separated in a purity greater than or equal to 90%.
The claims below are in addition to those above.
All refrences to claim(s) which appear below refer to the numbering after this setence.
1. A web page classification system, comprising:
a feature extractor that extracts at least one feature from a web page; and
a classifier that provides a classification of the web page based on the at least one extracted feature, the classification relating to whether the web page is a blog page.
2. The system of claim 1, wherein the classifier is a machine learning classifier.
3. The system of claim 1, wherein the at least one feature comprises the host of the web page.
4. The system of claim 1, wherein the at least one feature comprises at least one word or phrase contained in the web page.
5. The system of claim 1, wherein the at least one feature comprises the targets of outgoing links in the web page.
6. The system of claim 1, wherein the at least one feature comprises at least one string or substring in a uniform resource locator (URL) for the web page.
7. The system of claim 1, wherein the at least one feature comprises whether the web page contains an ATOM feed or an RSS feed.
8. The system of claim 1, wherein the at least one feature comprises a plurality of features selected from a set of features comprising at least one of the host of the web page, at least one word or phrase contained in the web page, the targets of outgoing links in the web page, at least one string or substring in a uniform resource locator (URL) for the web page, and whether the web page contains an ATOM feed or an RSS feed.
9. The system of claim 1, further comprising a web crawler that crawls a corpus of web pages and provides the web page to the feature extractor.
10. The system of claim 1, wherein the classification of the web page comprises an indication, prediction, or probability that the web page is a blog page or not.
11. A web page classification method, comprising:
extracting at least one feature from a web page using a feature extractor; and
classifying the web page as being a blog page or not based on the at least one extracted feature.
12. The method of claim 11, further comprising crawling a corpus of web pages and providing the web page to the feature extractor.
13. The method of claim 11, wherein the at least one feature is selected from a set of features comprising at least one of the host of the web page, at least one word or phrase contained in the web page, the targets of outgoing links in the web page, at least one string or substring in a uniform resource locator (URL) for the web page, and whether the web page contains an ATOM feed or an RSS feed.
14. The method of claim 11, wherein classifying the web page comprises providing an indication, prediction, or probability that the web page is a blog page or not.
15. The method of claim 14, further comprising:
determining if the probability is within a range, and if so, then further classifying the web page as being a blog page or not based on the at least one additional extracted feature.
16. The method of claim 11, further comprising:
classifying each of a plurality of additional web pages as being a blog page or not based on the at least one extracted feature;
forming a set of web pages that are classified as being a blog page; and
identifying a top level blog in the set of web pages.
17. A web page classification method, comprising:
classifying a plurality of web pages, each as being a blog page or not based on at least one extracted feature;
forming a set of web pages that are classified as being a blog page; and
identifying a top level blog in the set of web pages.
18. The method of claim 17, further comprising lexigraphically sorting the uniform resource locators (URLs) of each of the web pages in the set.
19. The method of claim 18, wherein identifying the top level blog comprises iterating through the lexigraphically sorted URLs to determine a common prefix of the web pages.
20. The method of claim 17, wherein the at least one extracted feature is selected from a set of features comprising at least one of the host of the web page, at least one word or phrase contained in the web page, the targets of outgoing links in the web page, at least one string or substring in a uniform resource locator (URL) for the web page, and whether the web page contains an ATOM feed or an RSS feed.