More info and prep for pagerank

This commit is contained in:
Ruben van de Ven 2020-01-31 19:44:17 +01:00
parent 2b48dd6c62
commit 4045f05121
11 changed files with 159 additions and 19 deletions

3
.gitmodules vendored Normal file
View file

@ -0,0 +1,3 @@
[submodule "ranking/PageRank"]
path = ranking/PageRank
url = https://github.com/ashkonf/PageRank

View file

@ -18,12 +18,16 @@ RewriteRule "^exhibition/mood_swings$" "index.html"
RewriteRule "^exhibition/route-du-nord$" "index.html"
RewriteRule "^exhibition/stateofemotion$" "index.html"
RewriteRule "^exhibition/the-new-current$" "index.html"
RewriteRule "^eye-without-a-face$" "index.html"
RewriteRule "^eye-without-a-face/video$" "index.html"
RewriteRule "^festival/statefestival$" "index.html"
RewriteRule "^guest-worker$" "index.html"
RewriteRule "^guest-worker/image/1$" "index.html"
RewriteRule "^guest-worker/image/2$" "index.html"
RewriteRule "^guest-worker/image/3$" "index.html"
RewriteRule "^guest-worker/video/1$" "index.html"
RewriteRule "^guest-worker/report$" "index.html"
RewriteRule "^guest-worker/video/outcome$" "index.html"
RewriteRule "^guest-worker/video/process$" "index.html"
RewriteRule "^heartbeat$" "index.html"
RewriteRule "^heartbeat/image/1$" "index.html"
RewriteRule "^in4art-salon$" "index.html"
@ -64,6 +68,7 @@ RewriteRule "^venue/tivoli$" "index.html"
RewriteRule "^venue/zoho$" "index.html"
RewriteRule "^workshop/plotting-data-das-theatre$" "index.html"
RewriteRule "^workshop/plotting-data-onomatopee$" "index.html"
RewriteRule "^www$" "index.html"
RewriteRule "^@type/Event$" "index.html"
RewriteRule "^@type/Report$" "index.html"
RewriteRule "^@type/VideoObject$" "index.html"
@ -79,3 +84,4 @@ RewriteRule "^@type/EventVenue$" "index.html"
RewriteRule "^@type/Museum$" "index.html"
RewriteRule "^@type/PerformingGroup$" "index.html"
RewriteRule "^@type/CollegeOrUniversity$" "index.html"
RewriteRule "^@type/WebSite$" "index.html"

BIN
assets/image/guest_worker02.jpg (Stored with Git LFS)

Binary file not shown.

File diff suppressed because one or more lines are too long

BIN
assets/thumb/guest_worker02.jpg (Stored with Git LFS)

Binary file not shown.

File diff suppressed because one or more lines are too long

2
ranking/.gitignore vendored Normal file
View file

@ -0,0 +1,2 @@
venv/

1
ranking/PageRank Submodule

@ -0,0 +1 @@
Subproject commit a138d0edb53135e0c070b3f5cc4a5b0b37dc778e

94
ranking/jsonld_ranking.py Normal file
View file

@ -0,0 +1,94 @@
from pyld import jsonld
import json
import PageRank.pagerank as pagerank
# use flattened jsonld
with open('../assets/js/rubenvandeven.jsonld', 'r') as fp:
contents = json.load(fp)['@graph']
linksPerItem = {}
def addCount(fromId, toId, weight):
if fromId not in linksPerItem:
linksPerItem[fromId] = {}
if not toId in linksPerItem[fromId]:
linksPerItem[fromId][toId] = 0
linksPerItem[fromId][toId] += weight
def addLink(fromId, toId, weight = 1):
addCount(fromId, toId, weight)
addCount(toId, fromId, weight)
for node in contents:
currentId = node['@id']
for key, value in node.items():
if type(value) == dict:
value = [value]
if type(value) == list:
for link in value:
if not type(link) == dict:
# can be any other list, eg. list of urls
continue
weight = 1
if key == 'https://schema.org/author':
weight = 2
elif key == 'https://schema.org/contributor':
weight = .5
# print(link)
linkedId = link['@id']
addLink(currentId, linkedId)
targetMin = .7
targetMax = 1.5
values = pagerank.powerIteration(linksPerItem)
normalised = targetMin + (values - min(values)) * (targetMax-targetMin)/max(values)
print(json.dumps(dict(normalised)))
#
# factor = 1 / min(values)
# print(factor)
# normalised = values * factor
# normalised -
#
# for(let nodeId in data) {
# let node = data[nodeId];
# let currentId = node["@id"];
# for(let key in node){
# let nodeAttr = Array.isArray(node[key]) ? node[key] : [node[key]];
# // // relations should always be lists (eases assumptions)
# // if(typeof node[key] !== "Array" && typeof node[key]['id'] !== "undefined") {
# // node[key] = [node[key]];
# // }
# // every attribute is an Array after flatten(), loop them
# for(let i in nodeAttr) {
# if(key !== "@id" && typeof nodeAttr[i] === "string" && nodes[nodeAttr[i]]) {
# links[links.length] = {
# "source": currentId,
# "target": nodeAttr[i],
# "name": key
# };
# }
# else if(typeof nodeAttr[i]["@id"] !== "undefined") {
# // if there is just one item, flatten/expand has turned urls in objects with just an id
# // reverse this, as we don't want these separate for this project
# if (Object.keys(nodeAttr[i]).length == 1 && typeof nodes[nodeAttr[i]["@id"]] === "undefined") {
# // skip
# // nodeAttr = nodeAttr[i]["id"];
# } else {
# links[links.length] = {
# "source": currentId,
# "target": nodeAttr[i]["@id"],
# "name": key
# };
# }
# }
# }
# }
# }

2
ranking/requirements.txt Normal file
View file

@ -0,0 +1,2 @@
PyLD==1.0.5
pandas

View file

@ -5,9 +5,9 @@
"r": "https://rubenvandeven.com"
},
"@type": "WebSite",
"@id": "r:#site",
"@id": "r:/www",
"url": "https://rubenvandeven.com",
"s:seeAlso": [ "https://rubenvandeven.com/rubenvandeven.jsonld", "https://rubenvandeven.com/foaf.rdf#me"],
"s:seeAlso": [ "https://rubenvandeven.com/rubenvandeven.jsonld", "https://git.rubenvandeven.com/r/portfolio"],
"author":
{
"@id": "r:/",
@ -199,14 +199,24 @@
"thumbnailUrl": "/assets\/thumb\/guest_worker03.jpg"
}
],
"video": {
"@id": "r:/guest-worker/video/1",
"@type": "VideoObject",
"thumbnailUrl": "/assets\/thumb\/guest_worker04.jpg",
"contentUrl": "http://works.rubenvandeven.com/2019-guest_worker/guestworker-without-credits.mp4" ,
"encodingFormat": "video/mp4",
"videoFrameSize": "1280x1024"
},
"video": [
{
"@id": "r:/guest-worker/video/outcome",
"@type": "VideoObject",
"thumbnailUrl": "/assets\/thumb\/guest_worker04.jpg",
"contentUrl": "http://works.rubenvandeven.com/2019-guest_worker/guestworker-without-credits.mp4" ,
"encodingFormat": "video/mp4",
"videoFrameSize": "1920x1080"
},
{
"@id": "r:/guest-worker/video/process",
"@type": "VideoObject",
"thumbnailUrl": "/assets\/thumb\/guest_worker05.jpg",
"contentUrl": "http://works.rubenvandeven.com/2019-guest_worker/GuestWorker.mp4" ,
"encodingFormat": "video/mp4",
"videoFrameSize": "1920x1080"
}
],
"author": {
"@id": "r:/person/merijn-van-moll",
"@type": "Person",
@ -273,7 +283,7 @@
},
"startDate": "2020-01-23",
"endDate": "2020-01-23",
"url": "https://ecp.nl/jaarcongres/kunst-expo-2019/",
"url": "https://www.tivolivredenburg.nl/agenda/privacyrede-2019-16-01-2019/",
"workFeatured": []
},
{
@ -284,7 +294,15 @@
"organizer": {"@id": "r:/organisation/ccu"},
"location": "Nijverheidsloods, Utrecht"
}
]
],
"about": {
"@id": "r:/guest-worker/report",
"@type": "Report",
"name": "Guest Worker Research Report",
"description": "A brief report written in light of Guest Worker (2019). In Dutch.",
"url": "http://guest_worker.rubenvandeven.com/",
"datePublished": "2019-11-26"
}
}
},
{
@ -824,6 +842,7 @@
]
},
{
"@id": "r:/eye-without-a-face",
"@type": "MediaObject",
"name": "EYE Without A Face",
"dateCreated": "2016",
@ -852,6 +871,14 @@
"thumbnailUrl": "/assets\/thumb\/eye_6-2.jpg"
}
],
"video": {
"@id": "r:/eye-without-a-face/video",
"@type": "VideoObject",
"thumbnailUrl": "/assets\/thumb\/eye_1.jpg",
"contentUrl": "http://works.rubenvandeven.com/2016-eye_without_a_face/EyeWithoutAFace.mp4" ,
"encodingFormat": "video/mp4",
"videoFrameSize": "1920x1080"
},
"@reverse": {
"workFeatured": [
{
@ -1323,7 +1350,7 @@
},
{
"@type": "TheaterEvent",
"name": "NEXT Festival",
"name": "Winzavod Art Center",
"location": "Moscow",
"startDate": "2020-02-14",
"endDate": "2020-02-28",