features.md

Features

Document features

Feature No.	Description
1	Stage0 score from TREC input file
2	BM25 Atire for the document
3	BM25 Atire for the document body
4	BM25 Atire for the document title
5	BM25 Atire for the document headings (h1-h4)
6	BM25 Atire for document inlinks
7	BM25 Atire for document a tags
8	BM25 TREC 3 (k = 1.2) for the document
9	BM25 TREC 3 (k = 1.2) for the document body
10	BM25 TREC 3 (k = 1.2) for the document title
11	BM25 TREC 3 (k = 1.2) for the document headings (h1-h4)
12	BM25 TREC 3 (k = 1.2) for the document inlinks
13	BM25 TREC 3 (k = 1.2) for the document a tags
14	BM25 TREC 3 (k = 2.0) for the document
15	BM25 TREC 3 (k = 2.0) for the document body
16	BM25 TREC 3 (k = 2.0) for the document title
17	BM25 TREC 3 (k = 2.0) for the document headings (h1-h4)
18	BM25 TREC 3 (k = 2.0) for the document inlinks
19	BM25 TREC 3 (k = 2.0) for the document a tags
20	LMDS (mu = 2500) for the document
21	LMDS (mu = 2500) for the document body
22	LMDS (mu = 2500) for the document title
23	LMDS (mu = 2500) for the document headings (h1-h4)
24	LMDS (mu = 2500) for the document inlinks
25	LMDS (mu = 2500) for the document a tags
26	LMDS (mu = 1500) for the document
27	LMDS (mu = 1500) for the document body
28	LMDS (mu = 1500) for the document title
29	LMDS (mu = 1500) for the document headings (h1-h4)
30	LMDS (mu = 1500) for the document inlinks
31	LMDS (mu = 1500) for the document a tags
32	LMDS (mu = 1000) for the document
33	LMDS (mu = 1000) for the document body
34	LMDS (mu = 1000) for the document title
35	LMDS (mu = 1000) for the document headings (h1-h4)
36	LMDS (mu = 1000) for the document inlinks
37	LMDS (mu = 1000) for the document a tags
38	TF.IDF for the document
39	TF.IDF for the document body
40	TF.IDF for the document title
41	TF.IDF for the document headings (h1-h4)
42	TF.IDF for the document inlinks
43	TF.IDF for the document a tags
44	Probability score for the document
45	Probability score for the document body
46	Probability score for the document title
47	Probability score for the document headings (h1-h4)
48	Probability score for the document inlinks
49	Probability score for the document a tags
50	Bose-Einstein score for the document
51	Bose-Einstein score for the document body
52	Bose-Einstein score for the document title
53	Bose-Einstein score for the document headings (h1-h4)
54	Bose-Einstein score for the document inlinks
55	Bose-Einstein score for the document a tags
56	DPH score for the document
57	DPH score for the document body
58	DPH score for the document title
59	DPH score for the document headings (h1-h4)
60	DPH score for the document inlinks
61	DPH score for the document a tags
62	DFR score for the document
63	DFR score for the document body
64	DFR score for the document title
65	DFR score for the document headings (h1-h4)
66	DFR score for the document inlinks
67	DFR score for the document a tags
68	Raw stream length of document
69	Raw stream length of document body
70	Raw stream length of document title
71	Raw stream length of document headings (h1-h4)
72	Raw stream length of document inlinks
73	Raw stream length of document a tags
74	Term frequency normalized sum of stream length of document
75	Term frequency normalized sum of stream length of document body
76	Term frequency normalized sum of stream length of document title
77	Term frequency normalized sum of stream length of document headings (h1-h4)
78	Term frequency normalized sum of stream length of document inlinks
79	Term frequency normalized sum of stream length of document a tags
80	Term frequency normalized min of stream length of document
81	Term frequency normalized min of stream length of document body
82	Term frequency normalized min of stream length of document title
83	Term frequency normalized min of stream length of document headings (h1-h4)
84	Term frequency normalized min of stream length of document inlinks
85	Term frequency normalized min of stream length of document a tags
86	Term frequency normalized max of stream length of document
87	Term frequency normalized max of stream length of document body
88	Term frequency normalized max of stream length of document title
89	Term frequency normalized max of stream length of document headings (h1-h4)
90	Term frequency normalized max of stream length of document inlinks
91	Term frequency normalized max of stream length of document a tags
92	Term frequency normalized mean of stream length of document
93	Term frequency normalized mean of stream length of document body
94	Term frequency normalized mean of stream length of document title
95	Term frequency normalized mean of stream length of document headings (h1-h4)
96	Term frequency normalized mean of stream length of document inlinks
97	Term frequency normalized mean of stream length of document a tags
98	Term frequency normalized variance of stream length of document
99	Term frequency normalized variance of stream length of document body
100	Term frequency normalized variance of stream length of document title
101	Term frequency normalized variance of stream length of document headings (h1-h4)
102	Term frequency normalized variance of stream length of document inlinks
103	Term frequency normalized variance of stream length of document a tags
104	TP-Score of document
105	Sum of BM25 query bigrams within an unordered window of 8 of the document
106	BM25 bigram interval score within a window of 100 of the document
107	SDM of document using default parameters
108	Frequency of query terms within the document title
109	Frequency of query terms within the document headings (h1-h4)
110	Frequency of query terms within the document
111	Frequency of query terms within the document inlinks
112	Document tag count for title
113	Document tag count for headings (h1-h4)
114	Document tag count for applet
115	Document tag count for object
116	Document tag count for embed

Static document features

Feature No.	Description
117	Document length
118	Title length
119	Visible term length
120	URL length
121	URL depth
122	Average term length
123	Entropy
124	Stop cover
125	Stop ratio
126	Fraction of anchor text
127	Fraction of visible text
128	Fraction of table text
129	Fraction of td text
130	Is wikipedia.org URL

Unigram pre-retrieval features

Feature No.	Description
131	Mean collection document frequency of query terms
132	Mean geometric mean of query terms
133	Min collection document frequency (query term)
134	Max collection document frequency (query term)
135	Min geometric mean (query term)
136	Max geometric mean (query term)
137	BM25 mean impact
138	BM25 mean mean
139	BM25 mean harmonic mean
140	BM25 mean median
141	BM25 mean interquartile range
142	BM25 mean variance
143	BM25 min
144	BM25 max
145	BM25 min mean
146	BM25 max mean
147	BM25 min median
148	BM25 max median
149	BM25 min harmonic mean
150	BM25 max harmonic mean
151	BM25 min variance
152	BM25 max variance
153	BM25 min first quartile
154	BM25 max first quartile
155	BM25 min third quartile
156	BM25 max third quartile
157	TF.IDF mean impact
158	TF.IDF mean mean
159	TF.IDF mean harmonic mean
160	TF.IDF mean median
161	TF.IDF mean interquartile range
162	TF.IDF mean variance
163	TF.IDF min
164	TF.IDF max
165	TF.IDF min mean
166	TF.IDF max mean
167	TF.IDF min median
168	TF.IDF max median
169	TF.IDF min harmonic mean
170	TF.IDF max harmonic mean
171	TF.IDF min variance
172	TF.IDF max variance
173	TF.IDF min first quartile
174	TF.IDF max first quartile
175	TF.IDF min third quartile
176	TF.IDF max third quartile
177	LM mean impact
178	LM mean mean
179	LM mean harmonic mean
180	LM mean median
181	LM mean interquartile range
182	LM mean variance
183	LM min
184	LM max
185	LM min mean
186	LM max mean
187	LM min median
188	LM max median
189	LM min harmonic mean
190	LM max harmonic mean
191	LM min variance
192	LM max variance
193	LM min first quartile
194	LM max first quartile
195	LM min third quartile
196	LM max third quartile
197	PR mean impact
198	PR mean mean
199	PR mean harmonic mean
200	PR mean median
201	PR mean interquartile range
202	PR mean variance
203	PR min
204	PR max
205	PR min mean
206	PR max mean
207	PR min median
208	PR max median
209	PR min harmonic mean
210	PR max harmonic mean
211	PR min variance
212	PR max variance
213	PR min first quartile
214	PR max first quartile
215	PR min third quartile
216	PR max third quartile
217	BE mean impact
218	BE mean mean
219	BE mean harmonic mean
220	BE mean median
221	BE mean interquartile range
222	BE mean variance
223	BE min
224	BE max
225	BE min mean
226	BE max mean
227	BE min median
228	BE max median
229	BE min harmonic mean
230	BE max harmonic mean
231	BE min variance
232	BE max variance
233	BE min first quartile
234	BE max first quartile
235	BE min third quartile
236	BE max third quartile
237	DPH mean impact
238	DPH mean mean
239	DPH mean harmonic mean
240	DPH mean median
241	DPH mean interquartile range
242	DPH mean variance
243	DPH min
244	DPH max
245	DPH min mean
246	DPH max mean
247	DPH min median
248	DPH max median
249	DPH min harmonic mean
250	DPH max harmonic mean
251	DPH min variance
252	DPH max variance
253	DPH min first quartile
254	DPH max first quartile
255	DPH min third quartile
256	DPH max third quartile
257	DFR mean impact
258	DFR mean mean
259	DFR mean harmonic mean
260	DFR mean median
261	DFR mean interquartile range
262	DFR mean variance
263	DFR min
264	DFR max
265	DFR min mean
266	DFR max mean
267	DFR min median
268	DFR max median
269	DFR min harmonic mean
270	DFR max harmonic mean
271	DFR min variance
272	DFR max variance
273	DFR min first quartile
274	DFR max first quartile
275	DFR min third quartile
276	DFR max third quartile

Query features

Feature No.	Description
277	Query length
278	Query length non-stop words
279	Simplified clarity score
280	Query scope score
281	TF.IDF average
282	TF.IDF variance
283	TF.IDF standard deviation
284	TF.IDF confidence
285	Gamma2: IDF max / IDF min
286	Average IDF on full query
287	Average IDF non-stop words
288	Average ICTF on full query
289	Average ICTF non-stop words