WIP

caliGraph.py

@@ -378,6 +378,9 @@ def scoreUnread(G, globMu, globStd):
         if node['t'] == 'book':
             if node['rating'] == None:
                 adjacens = list(G.adj[n].keys())
+                cats = {}
+                for cat in weights.keys():
+                    cats[cat] = []
                 for adj in adjacens:
                     adjNode = G.nodes[adj]
                     if 'score' in adjNode and adjNode['score'] != None:
@@ -385,6 +388,7 @@ def scoreUnread(G, globMu, globStd):
                         for fb in adjNode['feedbacks']:
                             feedbacks.append(fb)
                             ws.append(w)
+                            cats[adjNode['t']].append(fb)
                 if len(feedbacks):
                     node['mean'], node['std'] = norm.fit(feedbacks)
                     node['se'] = globStd / math.sqrt(len(feedbacks))
@@ -394,7 +398,18 @@ def scoreUnread(G, globMu, globStd):
                     ws.append(['se'])
                     feedbacks.append(globMu)
                     ws.append(['bias'])
-                    node['score'] = sum([fb*getWeightForType(w[0], w[1] if len(w)>1 else 1) for fb, w in zip(feedbacks, ws)])/sum([getWeightForType(w[0], w[1] if len(w)>1 else 1) for w in ws])
+                    linScore = sum([fb*getWeightForType(w[0], w[1] if len(w)>1 else 1) for fb, w in zip(feedbacks, ws)])/sum([getWeightForType(w[0], w[1] if len(w)>1 else 1) for w in ws])
+                    catWeightSum = sum([getWeightForType(c, 1) if cats[c] != [globMu] else 0 for c in weights])
+                    catScores = [(sum(cats[c])/len(cats[c])) * getWeightForType(c, 1) for c in cats if len(cats[c])]
+                    if len(catScores):
+                        normScore = sum(catScores)/len(catScores)
+                    else:
+                        normScore = globMu
+                    #print(catScores)
+                    #print(linScore, normScore)
+                    normF = getWeightForType('norm', 1)
+                    normF = 1
+                    node['score'] = linScore * (1-normF) + normScore * normF
                     node['_act'] = feedbacks
                     node['_wgh'] = ws
                 else:

neuralWeights.json

@@ -1 +1 @@
-{"topList": 0.8269026935523768, "recommender": 0.5687397308864482, "author": 0.9602300111040548, "series": 0.0015315822252870478, "tag": 0.0028789924118737056, "mu": 0.48730278196967397, "sigma": 0.02394878388858184, "se": 0.45207554705083647, "bias": 0.555294008129175}
+{"topList": 0.6813597954924836, "recommender": 0.7373820050627435, "author": 0.9322587113850848, "series": 0.9256656037821412, "tag": 0.9917141164721258, "mu": 0.5619155530016923, "sigma": 0.05411831944351521, "se": 0.6823826056339866, "bias": 0.7987740969341441, "norm": 0.9443733783393993}