Class: Entry
- Inherits:
-
ActiveRecord::Base
- Object
- ActiveRecord::Base
- Entry
- Defined in:
- app/models/entry.rb
Overview
Schema Information
Table name: entries
id :integer(4) not null, primary key
feed_id :integer(4) not null
permalink :string(2083) default(""), not null
author :string(2083)
title :text default(""), not null
description :text
content :text
unique_content :boolean(1)
published_at :datetime not null
entry_updated_at :datetime
harvested_at :datetime
oai_identifier :string(2083)
language_id :integer(4)
direct_link :string(2083)
indexed_at :datetime default(Fri Jan 01 01:01:01 UTC 1971), not null
relevance_calculated_at :datetime default(Fri Jan 01 01:01:01 UTC 1971), not null
popular :text
relevant :text
other :text
grain_size :string(255) default("unknown")
Indexes
index_entries_on_direct_link (direct_link)
index_entries_on_feed_id (feed_id)
index_entries_on_indexed_at (indexed_at)
index_entries_on_language_id (language_id)
index_entries_on_oai_identifier (oai_identifier)
index_entries_on_permalink (permalink)
index_entries_on_published_at (published_at)
index_entries_on_relevance_calculated_at (relevance_calculated_at)
index_entries_on_grain_size (grain_size)
Constant Summary collapse
- @@default_time_on_page =
60.0
Class Method Summary collapse
- .avg_time(clicks, old_avg, time_on_page) ⇒ Object
- .recommender_entry(uri) ⇒ Object
- .search(search_terms, grain_size = nil, language = "en", limit = 10, offset = 0, operator = :or) ⇒ Object
- .top_tags(tags = nil) ⇒ Object
- .track_click(session, recommendation_id, referrer, redirect_type = "direct_link", requester = "unknown", user_agent = "unknown") ⇒ Object
- .track_time_on_page(session, uri) ⇒ Object
- .truncate_words(text, length = 30, end_string = ' ...') ⇒ Object
Instance Method Summary collapse
- #calc_click_threshold(recs) ⇒ Object
- #calc_relevance_threshold(recs) ⇒ Object
- #json_recommendations(limit = 5, order = "mixed", details = false, omit_feeds = nil) ⇒ Object
- #randomize(recs) ⇒ Object
- #rank_recommendations ⇒ Object
- #ranked_recommendations(limit = 5, order = "mixed", details = false, omit_feeds = nil) ⇒ Object
- #recommendation_entries(limit = 20, order = "relevance", details = false, omit_feeds = nil) ⇒ Object
- #recommendations(limit = 20, order = "relevance", details = false, omit_feeds = nil) ⇒ Object
- #relevant_recommendations(limit = 5, order = "relevance", details = false, omit_feeds = nil) ⇒ Object
- #relevant_recommendations_filtered(limit, details, omit_feeds) ⇒ Object
- #resource_uri ⇒ Object
Class Method Details
.avg_time(clicks, old_avg, time_on_page) ⇒ Object
120 121 122 |
# File 'app/models/entry.rb', line 120 def self.avg_time(clicks, old_avg, time_on_page) return (old_avg*(clicks-1) + time_on_page)/clicks end |
.recommender_entry(uri) ⇒ Object
67 68 69 70 |
# File 'app/models/entry.rb', line 67 def self.recommender_entry(uri) uri = normalized_uri(uri) Entry.find(:first, :conditions => ['permalink = ? OR direct_link = ?', uri, uri], :order => 'direct_link IS NULL DESC') || Entry.new(:permalink => uri) end |
.search(search_terms, grain_size = nil, language = "en", limit = 10, offset = 0, operator = :or) ⇒ Object
62 63 64 65 |
# File 'app/models/entry.rb', line 62 def self.search(search_terms, grain_size = nil, language = "en", limit = 10, offset = 0, operator = :or) query = (!grain_size.nil? && grain_size != 'all') ? (search_terms + ") AND (grain_size:#{grain_size}") : search_terms return find_by_solr(query, :limit => limit, :offset => offset, :scores => true, :select => "entries.id, entries.title, entries.permalink, entries.direct_link, entries.published_at, entries.description, entries.feed_id, feeds.short_title AS collection", :joins => "INNER JOIN feeds ON feeds.id = entries.feed_id", :core => language, :operator => operator) end |
.top_tags(tags = nil) ⇒ Object
58 59 60 |
# File 'app/models/entry.rb', line 58 def self.( = nil) end |
.track_click(session, recommendation_id, referrer, redirect_type = "direct_link", requester = "unknown", user_agent = "unknown") ⇒ Object
124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 |
# File 'app/models/entry.rb', line 124 def self.track_click(session, recommendation_id, referrer, redirect_type = "direct_link", requester = "unknown", user_agent = "unknown") # look up the recommendation recommendation = Recommendation.find(recommendation_id) return "" if !recommendation # get the entries being linked from and to entry = Entry.find(recommendation.entry_id) target = Entry.find(recommendation.dest_entry_id) # get the list of recommendations that have been clicked during this session clicks = session[:rids] || Array.new # redirect to our frame page redirect = "/visits/#{recommendation.dest_entry_id}" # track the time on the last page track_time_on_page(session, redirect) # if this is first time the user clicked on this recommendation during this session if !clicks.include?(recommendation_id) # add this recommendation to the end of the list clicks << recommendation_id session[:rids] = clicks # update the click time recommendation.avg_time_at_dest = ((recommendation.avg_time_at_dest*recommendation.clicks) + @@default_time_on_page)/(recommendation.clicks + 1) recommendation.clicks += 1 recommendation.save! # store info about this click in the session now = Time.now session[:last_clicked_recommendation] = recommendation_id session[:last_clicked_recommendation_time] = now session[:last_clicked_recommendation_uri] = redirect # update the recommendation cache for the entry entry.rank_recommendations if entry # track the click in the db Click.create(:recommendation_id => recommendation_id, :when => now, :referrer => referrer, :requester => requester, :user_agent => user_agent) end return redirect end |
.track_time_on_page(session, uri) ⇒ Object
98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 |
# File 'app/models/entry.rb', line 98 def self.track_time_on_page(session, uri) recommendation_id = session[:last_clicked_recommendation] if !recommendation_id.nil? time_on_page = (Time.now - session[:last_clicked_recommendation_time].to_f).to_i # if they spend longer than two minutes on a page, we don't infer anything if time_on_page > 5 and time_on_page < 120 if normalized_uri(uri) != session[:last_clicked_recommendation_uri] recommendation = Recommendation.find(recommendation_id) entry = Entry.find(recommendation.entry_id) new_avg = (recommendation.avg_time_at_dest*recommendation.clicks - @@default_time_on_page + time_on_page)/recommendation.clicks recommendation.avg_time_at_dest = new_avg recommendation.save! entry.rank_recommendations session[:last_clicked_recommendation] = nil end else session[:last_clicked_recommendation] = nil if time_on_page > 5 end end end |
.truncate_words(text, length = 30, end_string = ' ...') ⇒ Object
174 175 176 177 |
# File 'app/models/entry.rb', line 174 def self.truncate_words(text, length = 30, end_string = ' ...') words = text.split() words[0..(length-1)].join(' ') + (words.length > length ? end_string : '') end |
Instance Method Details
#calc_click_threshold(recs) ⇒ Object
302 303 304 305 306 307 308 309 310 311 312 313 314 315 |
# File 'app/models/entry.rb', line 302 def calc_click_threshold(recs) sum = 0 recs.each do |r| sum += r["clicks"].to_f end average = sum/recs.length sum = 0 recs.each do |r| sum += (r["clicks"].to_f-average)**2 end standard_deviation = Math.sqrt(sum/recs.length); threshold = average + standard_deviation return threshold > 5 ? threshold : 5 end |
#calc_relevance_threshold(recs) ⇒ Object
288 289 290 291 292 293 294 295 296 297 298 299 300 |
# File 'app/models/entry.rb', line 288 def calc_relevance_threshold(recs) sum = 0 recs.each do |r| sum += r["relevance"].to_f end average = sum/recs.length sum = 0 recs.each do |r| sum += (r["relevance"].to_f-average)**2 end standard_deviation = Math.sqrt(sum/recs.length); return average + standard_deviation end |
#json_recommendations(limit = 5, order = "mixed", details = false, omit_feeds = nil) ⇒ Object
235 236 237 |
# File 'app/models/entry.rb', line 235 def json_recommendations(limit = 5, order = "mixed", details = false, omit_feeds = nil) ActiveSupport::JSON.encode(ranked_recommendations(limit, order, details, omit_feeds)) end |
#randomize(recs) ⇒ Object
274 275 276 277 278 279 280 281 282 283 284 285 286 |
# File 'app/models/entry.rb', line 274 def randomize(recs) i = recs.length return recs if (i == 0) while (i > 0) i = i - 1 j = (rand*(i+1)).floor ti = recs[i] tj =recs[j] recs[i] = tj recs[j] = ti end return recs end |
#rank_recommendations ⇒ Object
239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 |
# File 'app/models/entry.rb', line 239 def rank_recommendations return # get recommendations for the entry from the recommendations table recs = self.recommendations # for storing the various lists popular_recs = [] relevant_recs = [] other_recs = [] # see where to cut draw the lines for popular and relevant click_threshold = calc_click_threshold(recs) relevance_threshold = calc_relevance_threshold(recs) # store the recommendations recs.each do |r| if (r["clicks"].to_i > click_threshold) popular_recs << r elsif (r["relevance"].to_f > relevance_threshold) relevant_recs << r else other_recs << r end end # order popular items strictly by clicks popular_recs.sort{|r1,r2| r2["avg_time_on_target"].to_i <=> r1["avg_time_on_target"].to_i} # cache the JSON for the lists in the entry record self.popular = ActiveSupport::JSON.encode(popular_recs) self.relevant = ActiveSupport::JSON.encode(relevant_recs) self.other = ActiveSupport::JSON.encode(other_recs) self.save! end |
#ranked_recommendations(limit = 5, order = "mixed", details = false, omit_feeds = nil) ⇒ Object
179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 |
# File 'app/models/entry.rb', line 179 def ranked_recommendations(limit = 5, order = "mixed", details = false, omit_feeds = nil) if self.id.nil? return Entry.real_time_recommendations(self.permalink, details, :core => 'en', :limit => limit) else return relevant_recommendations(limit, "clicks DESC, relevance", details, omit_feeds) if order == "clicks" return relevant_recommendations(limit, "relevance", details, omit_feeds) if (order == "relevance" || details == true) return relevant_recommendations_filtered(limit, details, omit_feeds) if omit_feeds != nil end recs = [] if self.popular != nil && !self.popular.empty? recs.concat(ActiveSupport::JSON.decode(self.popular).first(limit)) end if recs.length < limit && self.relevant != nil && !self.relevant.empty? relevant_recs = randomize(ActiveSupport::JSON.decode(self.relevant)) recs.concat(relevant_recs.first(limit - recs.length)) end if recs.length < limit && self.relevant != nil && !self.other.empty? other_recs = randomize(ActiveSupport::JSON.decode(self.other)) recs.concat(other_recs.first(limit - recs.length)) end return recs end |
#recommendation_entries(limit = 20, order = "relevance", details = false, omit_feeds = nil) ⇒ Object
85 86 87 88 89 90 91 92 93 94 95 96 |
# File 'app/models/entry.rb', line 85 def recommendation_entries(limit = 20, order = "relevance", details = false, omit_feeds = nil) sql = "SELECT recommendations.dest_entry_id AS id, entries.permalink, entries.title, entries.description, entries.direct_link, feeds.short_title AS collection " sql << ", relevance_calculated_at, relevance, clicks, avg_time_at_dest AS avg_time_on_target, author, published_at " if details == true sql << "FROM recommendations " sql << "INNER JOIN entries ON recommendations.dest_entry_id = entries.id " sql << "INNER JOIN feeds ON entries.feed_id = feeds.id " sql << "WHERE recommendations.entry_id = ? " sql << ("AND entries.feed_id NOT IN (" + omit_feeds.gsub(/[^0-9,]/,'') + ") ") if omit_feeds != nil sql << "ORDER BY " + order + " DESC " sql << "LIMIT " + limit.to_s Entry.find_by_sql([sql,self.id]) end |
#recommendations(limit = 20, order = "relevance", details = false, omit_feeds = nil) ⇒ Object
72 73 74 75 76 77 78 79 80 81 82 83 |
# File 'app/models/entry.rb', line 72 def recommendations(limit = 20, order = "relevance", details = false, omit_feeds = nil) sql = "SELECT recommendations.id, dest_entry_id, entries.permalink, entries.title, entries.description, entries.direct_link, feeds.short_title AS collection " sql << ", relevance_calculated_at, relevance, clicks, avg_time_at_dest AS avg_time_on_target, author, published_at " if details == true sql << "FROM recommendations " sql << "INNER JOIN entries ON recommendations.dest_entry_id = entries.id " sql << "INNER JOIN feeds ON entries.feed_id = feeds.id " sql << "WHERE recommendations.entry_id = ? " sql << ("AND entries.feed_id NOT IN (" + omit_feeds.gsub(/[^0-9,]/,'') + ") ") if omit_feeds != nil sql << "ORDER BY " + order + " DESC " sql << "LIMIT " + limit.to_s Entry.find_by_sql([sql,self.id]) end |
#relevant_recommendations(limit = 5, order = "relevance", details = false, omit_feeds = nil) ⇒ Object
170 171 172 |
# File 'app/models/entry.rb', line 170 def relevant_recommendations(limit = 5, order = "relevance", details = false, omit_feeds = nil) return self.recommendations(limit, order, details, omit_feeds) end |
#relevant_recommendations_filtered(limit, details, omit_feeds) ⇒ Object
203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 |
# File 'app/models/entry.rb', line 203 def relevant_recommendations_filtered(limit, details, omit_feeds) # get recommendations for the entry from the recommendations table recs = self.recommendations(limit, "mixed", details, omit_feeds) # for storing the various lists popular_recs = [] relevant_recs = [] other_recs = [] # see where to cut draw the lines for popular and relevant click_threshold = calc_click_threshold(recs) relevance_threshold = calc_relevance_threshold(recs) # store the recommendations recs.each do |r| if (r["clicks"].to_i > click_threshold) popular_recs << r elsif (r["relevance"].to_f > relevance_threshold) relevant_recs << r else other_recs << r end end # order popular items strictly by clicks popular_recs.sort{|r1,r2| r2["avg_time_on_target"].to_i <=> r1["avg_time_on_target"].to_i} return popular_recs[0..limit] if popular_recs.size > limit return (popular_recs + relevant_recs)[0..limit] if (popular_recs.size + relevant_recs.size) > limit return (popular_recs + relevant_recs + other_recs)[0..limit] end |
#resource_uri ⇒ Object
54 55 56 |
# File 'app/models/entry.rb', line 54 def resource_uri self.direct_link.nil? ? self.permalink : self.direct_link end |