Semantic Similarity¶

This module contains the classes for the evaluation of the Semantic Similarity. Please refer to the single classes for details on the implemented measures.

SemSim.TermSemSim¶

This class provides the prototype for Term semantic similarity measures (TSS)

There are two types of Term semantic similarity: a first group that can evaluate the semantic similarity between two sets of terms (groupwise - G_TSS), and a second group that can only evaluate the similarity between pairs of GO terms (pairwise - P_TSS). Each class extending TermSemSim should declare whether it is groupwise or pairwise.

TermSemSim relies on SemSimUtils to perform a lot of tasks (e.g. evaluating Term IC or common ancestors). A SemSimUtils object can be passed to the constructor as input data. Otherwise, a new instance will be created. Using only one copy of SemSimUtils helps reducing time and space requirements and is strongly recommended.

exception fastsemsim.SemSim.TermSemSim.MissingAcException(message)¶: Bases: exceptions.Exception

class fastsemsim.SemSim.TermSemSim.TermSemSim(ontology, ac=None, util=None, do_log=False)¶

Bases: object

G_TSS = 'Groupwise'¶

IC_based = None¶

P_TSS = 'Pairwise'¶

SS_type = None¶

SemSim(term1, term2, ontology=None)¶

format_and_check_data = True¶

setSanityCheck(en)¶

SemSim.ObjSemSim¶

This class provides the prototype for a generic Object Semantic Similarity measure

class fastsemsim.SemSim.ObjSemSim.ObjSemSim(ontology, ac, TSS=None, MSS=None, util=None, do_log=False)¶

Bases: object

SemSim(obj1, obj2, root=None)¶

SemSim.ObjSetSemSim¶

This class provides the prototype for a generic Object Set Semantic Similarity measure (PSS)

class fastsemsim.SemSim.ObjSetSemSim.ObjSetSemSim(ontology, ac, TSS=None, MSS=None, util=None, do_log=False)¶

SemSim(obj1, obj2, root=None)¶

SemSim.SetSemSim¶

This class provides the prototype for a generic Pairwise Object Semantic Similarity measure

class fastsemsim.SemSim.SetSemSim.SetSemSim(ontology, ac=None, TSS=None, MSS=None, util=None, do_log=False)¶

SemSim(obj1, obj2, root=None)¶

Specific Semantic Similarity measures¶

SemSim.ResnikSemSim¶

Resnik Semantic Similarity Measure

Reference: Resnik, P. (1999). Semantic similarity in a taxonomy: An information-based measure and its application to problems of ambiguity in natural language. Journal of Artificial Intelligence Research, 11, 95-130.

class fastsemsim.SemSim.ResnikSemSim.ResnikSemSim(ontology, ac=None, util=None, do_log=False)¶

Bases: fastsemsim.SemSim.TermSemSim.TermSemSim

IC_based = True¶

SS_type = 'Pairwise'¶

SemSim.CosineSemSim¶

Cosine Semantic Similarity Measure

Reference:

class fastsemsim.SemSim.CosineSemSim.CosineSemSim(ontology, ac=None, util=None, do_log=False)¶

Bases: fastsemsim.SemSim.TermSemSim.TermSemSim

IC_based = False¶

SS_type = 'Groupwise'¶

dotprod(vector1, vector2)¶

extend_annotations = True¶

SemSim.CzekanowskiDiceSemSim¶

Czekanowski and Dice Semantic Similarity Measure

Reference:

class fastsemsim.SemSim.CzekanowskiDiceSemSim.CzekanowskiDiceSemSim(ontology, ac=None, util=None, do_log=False)¶

Bases: fastsemsim.SemSim.TermSemSim.TermSemSim

IC_based = False¶

SS_type = 'Groupwise'¶

extend_annotations = True¶

SemSim.DiceSemSim¶

Dice Semantic Similarity Measure

Reference:

class fastsemsim.SemSim.DiceSemSim.DiceSemSim(ontology, ac=None, util=None, do_log=False)¶

Bases: fastsemsim.SemSim.TermSemSim.TermSemSim

IC_based = False¶

SS_type = 'Groupwise'¶

extend_annotations = True¶

SemSim.GSESAMESemSim¶

G-SESAME Semantic Similarity Measure

Reference:

class fastsemsim.SemSim.GSESAMESemSim.GSESAMESemSim(ontology, ac=None, util=None, do_log=False)¶

Bases: fastsemsim.SemSim.TermSemSim.TermSemSim

IC_based = False¶

SS_type = 'Pairwise'¶

generic_score = 0.5¶

is_a_score = 0.8¶

neg_regulates_score = 0.6¶

part_of_score = 0.6¶

pos_regulates_score = 0.6¶

regulates_score = 0.6¶

score_ancestors(term)¶

score_edge(tp, t)¶

SemSim.JaccardSemSim¶

Jaccard Index based Semantic Similarity Measure

Reference:

class fastsemsim.SemSim.JaccardSemSim.JaccardSemSim(ontology, ac=None, util=None, do_log=False)¶

Bases: fastsemsim.SemSim.TermSemSim.TermSemSim

IC_based = False¶

SS_type = 'Groupwise'¶

extend_annotations = True¶

SemSim.JiangConrathSemSim¶

Jiang and Conrath Semantic Similarity Measure

Reference:

class fastsemsim.SemSim.JiangConrathSemSim.JiangConrathSemSim(ontology, ac=None, util=None, do_log=False)¶

Bases: fastsemsim.SemSim.TermSemSim.TermSemSim

IC_based = True¶

SS_type = 'Pairwise'¶

SemSim.LinSemSim¶

Lin Semantic Similarity Measure

Reference:

class fastsemsim.SemSim.LinSemSim.LinSemSim(ontology, ac=None, util=None, do_log=False)¶

Bases: fastsemsim.SemSim.TermSemSim.TermSemSim

IC_based = True¶

SS_type = 'Pairwise'¶

SemSim.SimGICSemSim¶

SimGIC Semantic Similarity Measure

Reference:

class fastsemsim.SemSim.SimGICSemSim.SimGICSemSim(ontology, ac=None, util=None, do_log=False)¶

Bases: fastsemsim.SemSim.TermSemSim.TermSemSim

IC_based = True¶

SS_type = 'Groupwise'¶

SemSim.SimICNDSemSim¶

ICND Semantic Similarity Measure

Reference:

class fastsemsim.SemSim.SimICNDSemSim.ICNDSemSim(ontology, ac=None, util=None, do_log=False)¶

Bases: fastsemsim.SemSim.TermSemSim.TermSemSim

IC_based = True¶

SS_type = 'Pairwise'¶

generic_score = 1.0¶

is_a_score = 1.0¶

neg_regulates_score = 1.0¶

part_of_score = 1.0¶

pos_regulates_score = 1.0¶

regulates_score = 1.0¶

score_ancestors(term)¶

score_edge(tp, t)¶

SemSim.SimICNPSemSim¶

ICNP Semantic Similarity Measure

Reference:

class fastsemsim.SemSim.SimICNPSemSim.ICNPSemSim(ontology, ac=None, util=None, do_log=False)¶

Bases: fastsemsim.SemSim.TermSemSim.TermSemSim

IC_based = True¶

SS_type = 'Pairwise'¶

generic_score = 1.0¶

is_a_score = 1.0¶

neg_regulates_score = 1.0¶

part_of_score = 1.0¶

pos_regulates_score = 1.0¶

regulates_score = 1.0¶

score_ancestors(term)¶

score_edge(tp, t)¶

SemSim.SimICSemSim¶

Information Content Semantic Similarity Measure

Reference:

class fastsemsim.SemSim.SimICSemSim.SimICSemSim(ontology, ac=None, util=None, do_log=False)¶

Bases: fastsemsim.SemSim.TermSemSim.TermSemSim

IC_based = True¶

SS_type = 'Pairwise'¶

use_Lin = True¶

SemSim.SimNTOSemSim¶

Normalized Term Overlap Semantic Similarity Measure

Reference:

class fastsemsim.SemSim.SimNTOSemSim.SimNTOSemSim(ontology, ac=None, util=None, do_log=False)¶

Bases: fastsemsim.SemSim.TermSemSim.TermSemSim

IC_based = False¶

SS_type = 'Groupwise'¶

extend_annotations = True¶

SemSim.SimRelSemSim¶

SimRel Semantic Similarity Measure

Reference:

class fastsemsim.SemSim.SimRelSemSim.SimRelSemSim(ontology, ac=None, util=None, do_log=False)¶

Bases: fastsemsim.SemSim.TermSemSim.TermSemSim

IC_based = True¶

SS_type = 'Pairwise'¶

use_Lin = True¶

SemSim.SimTOSemSim¶

SimTO Semantic Similarity Measure

Reference:

class fastsemsim.SemSim.SimTOSemSim.SimTOSemSim(ontology, ac=None, util=None, do_log=False)¶

Bases: fastsemsim.SemSim.TermSemSim.TermSemSim

IC_based = False¶

SS_type = 'Groupwise'¶

SemSim.SimUISemSim¶

SimUI Semantic Similarity Measure

Reference:

class fastsemsim.SemSim.SimUISemSim.SimUISemSim(ontology, ac=None, util=None, do_log=False)¶

Bases: fastsemsim.SemSim.TermSemSim.TermSemSim

IC_based = False¶

SS_type = 'Groupwise'¶

SemSim.MixSemSim¶

This class provides the prototype for a generic mixing strategy for pairwise Term Semantic Similarity measures

class fastsemsim.SemSim.MixSemSim.MixSemSim(ontology, ac, util=None, do_log=False)¶

Bases: object

SemSim(set1, set2, TSS)¶

SemSim.avgSemSim¶

This class defines the prototype for a generic mixing strategy for pairwise term Protein Semantic Similarity measures

class fastsemsim.SemSim.avgSemSim.avgSemSim(ontology, ac, util=None, do_log=False)¶: Bases: fastsemsim.SemSim.MixSemSim.MixSemSim

SemSim.maxSemSim¶

Max mixing strategy

class fastsemsim.SemSim.maxSemSim.maxSemSim(ontology, ac, util=None, do_log=False)¶: Bases: fastsemsim.SemSim.MixSemSim.MixSemSim

SemSim.BMASemSim¶

Best Match Average (BMA) mixing strategy for pairwise term Protein Semantic Similarity measures

class fastsemsim.SemSim.BMASemSim.BMASemSim(ontology, ac, util=None, do_log=False)¶

Bases: fastsemsim.SemSim.MixSemSim.MixSemSim

fair = True¶

SemSim.SemSimUtils¶

This class provides some routines to calculate basic properties used by different SS measures. In particular this class provides code for evaluating:

term ICs
term frequency within an annotation corpus
term’s ancestors
term’s offspring
terms’s children
terms’s parents
MICA/DCA/LCA
term’s distance

class fastsemsim.SemSim.SemSimUtils.SemSimUtils(ontology, ac=None)¶

Bases: object

det_IC(term)¶

det_IC_table()¶

det_MICA(term1, term2)¶

det_ancestors_union(term1, term2)¶

det_common_ancestors(term1, term2)¶

difference(set1, set2)¶

get_ancestors(term1)¶

int_det_IC(term_id)¶

int_det_IC_table()¶

int_det_ancestors(goid, temp_intra)¶

int_det_ancestors_table()¶

int_det_freq(term_id)¶

int_det_freq_table()¶

int_det_lineage()¶

int_det_offspring(goid, temp_intra)¶

int_det_offspring_table()¶

int_det_p(term_id)¶

int_det_p_table()¶

int_merge_sets(set1, set2)¶

intersection(set1, set2)¶

Semantic Similarity¶

SemSim.TermSemSim¶

SemSim.ObjSemSim¶

SemSim.ObjSetSemSim¶

SemSim.SetSemSim¶

Specific Semantic Similarity measures¶

SemSim.ResnikSemSim¶

SemSim.CosineSemSim¶

SemSim.CzekanowskiDiceSemSim¶

SemSim.DiceSemSim¶

SemSim.GSESAMESemSim¶

SemSim.JaccardSemSim¶

SemSim.JiangConrathSemSim¶

SemSim.LinSemSim¶

SemSim.SimGICSemSim¶

SemSim.SimICNDSemSim¶

SemSim.SimICNPSemSim¶

SemSim.SimICSemSim¶

SemSim.SimNTOSemSim¶

SemSim.SimRelSemSim¶

SemSim.SimTOSemSim¶

SemSim.SimUISemSim¶

SemSim.MixSemSim¶

SemSim.avgSemSim¶

SemSim.maxSemSim¶

SemSim.BMASemSim¶

SemSim.SemSimUtils¶

Table Of Contents

Previous topic

Next topic

This Page

Navigation

Semantic Similarity¶

SemSim.TermSemSim¶

SemSim.ObjSemSim¶

SemSim.ObjSetSemSim¶

SemSim.SetSemSim¶

Specific Semantic Similarity measures¶

SemSim.ResnikSemSim¶

SemSim.CosineSemSim¶

SemSim.CzekanowskiDiceSemSim¶

SemSim.DiceSemSim¶

SemSim.GSESAMESemSim¶

SemSim.JaccardSemSim¶

SemSim.JiangConrathSemSim¶

SemSim.LinSemSim¶

SemSim.SimGICSemSim¶

SemSim.SimICNDSemSim¶

SemSim.SimICNPSemSim¶

SemSim.SimICSemSim¶

SemSim.SimNTOSemSim¶

SemSim.SimRelSemSim¶

SemSim.SimTOSemSim¶

SemSim.SimUISemSim¶

SemSim.MixSemSim¶

SemSim.avgSemSim¶

SemSim.maxSemSim¶

SemSim.BMASemSim¶

SemSim.SemSimUtils¶

Table Of Contents

Previous topic

Next topic

This Page

Quick search

Navigation