OfR|SrSSKrSSKrSSKrSSKJr SSKrSSKJ r "SS5r "SS5r "S S 5r g) a If you use the VADER sentiment analysis tools, please cite: Hutto, C.J. & Gilbert, E.E. (2014). VADER: A Parsimonious Rule-based Model for Sentiment Analysis of Social Media Text. Eighth International Conference on Weblogs and Social Media (ICWSM-14). Ann Arbor, MI, June 2014. N)product)pairwisec.\rSrSrSrSrSrSrSr1Skr 0S\_S \_S \_S \_S \_S \_S\_S\_S\_S\_S\_S\_S\_S\_S\_S\_S\_0S\_S\_S\_S\_S\_S\_S\_S \_S!\_S"\_S#\_S$\_S%\_S&\_S'\_S(\_S)\_E0S*\_S+\_S,\_S-\_S.\_S/\_S0\_S1\_S2\_S3\_S4\_S5\_S6\_S7\_S8\_S9\_S:\_E\\\\\\\\\\\\\\\S;.Er SS?S@S>SA.r \ R"SB\ R"\R 5SC35r/SDQrSErSKSFjrSLSGjrSHrSIrgJ)MVaderConstants!z0 A class to keep the Vader lists and constants. gn?gnҿg~jt?gGz>;nornotaintcantdontisntnonenopeuhuhwontarentdidnthadnthasntnevershantwasntcannotdarentdoesnthaventmustntneedntrarelyseldomwerentcouldntdespitemightntneithernothingnowhereoughtntwithoutwouldntshouldntain'tcan'tdon'tisn'tuh-uhwon'taren'tdidn'thadn'thasn'tshan'twasn'tdaren'tdoesn'thaven'tmustn'tneedn'tweren'tcouldn'tmightn'toughtn'twouldn't shouldn't absolutely amazinglyawfully completely considerably decidedlydeeplyeffing enormouslyentirely especially exceptionally extremely fabulouslyflippingflippinfrickingfrickinfriggingfrigginfullyfuckinggreatlyhellahighlyhugely incredibly intenselymajorlymoremost particularlypurelyquitereally remarkablyso substantially thoroughlytotally tremendouslyuber unbelievably unusuallyutterlyveryalmostbarelyhardlyz just enoughzkind of)kindakindofzkind-oflesslittle marginally occasionallypartlyscarcelyslightlysomewhatzsort ofsortasortofzsort-of?g)zthe shitzthe bombzbad assz yeah rightzcut the mustardz kiss of deathz hand to mouth[]).!?,;:-'"z!!z!!!z??z???z?!?z!?!z?!?!z!?!?cgN)selfs 6/usr/lib/python3/dist-packages/nltk/sentiment/vader.py__init__VaderConstants.__init__s c^URm[U4SjU55(agU(a[SU55(ag[U5H1up4UR5S:XdMUR5S:wdM1 g g)z, Determine if input contains negation words c3H># UHoR5T;v M g7frlower).0word neg_wordss r )VaderConstants.negated..sA[Tzz|y([s"Tc3H# UHnSUR5;v M g7f)zn'tNr)rrs rrrsA[T5DJJL([s "leastatF)NEGATEanyrr)r input_words include_ntfirstsecondrs @rnegatedVaderConstants.negatedsmKK A[A A A A[AAA%k2ME||~(U[[]d-B3rcBU[R"X-U-5- nU$)zd Normalize the score to be between -1 and 1 using an alpha that approximates the max expected value )mathsqrt)rscorealpha norm_scores r normalizeVaderConstants.normalizes$ TYY '>?? rcSnUR5nXPR;aZURUnUS:aUS-nUR5(a+U(a$US:aX@R- nU$X@R-nU$)zP Check if the preceding words increase, decrease, or negate/nullify the valence r)r BOOSTER_DICTisupperC_INCR)rrvalence is_cap_diffscalar word_lowers rscalar_inc_decVaderConstants.scalar_inc_decsz ZZ\ ** *&&z2F{" ||~~+Q;kk)F kk)F rrN)T))__name__ __module__ __qualname____firstlineno____doc__B_INCRB_DECRrN_SCALARrrSPECIAL_CASE_IDIOMSrecompileescapestring punctuationREGEX_REMOVE_PUNCTUATION PUNC_LISTrrrr__static_attributes__rrrrr!s F FFH<FBCfCVC 6C f C  C V C &C &C fC FC fC C VC fC FC 6!C" F#C$ 6%C& F'C( 6)C* +C, 6-C. 6/C0 1C2 &3C4 &5C6 f7C8 V9C: 6;C< =C> ?C@ ACB &CCD ECF &GCH fICJ fKCL MCN fOCP 6QCR SCT UCV WCX VYCZ 6[C\ ]C^ &_C` &aCb &cCd veCf 6gChECLN "zzAbii8J8J.K-LA*NOI(  rrc0\rSrSrSrSrSrSrSrSr g) SentiTextizD Identify sentiment-relevant string-level properties of input text. c[U[5(d[URS55nXlX lX0lUR 5UlURUR5Ul g)Nzutf-8) isinstancestrencodetextrr_words_and_emoticonswords_and_emoticonsallcap_differentialr)rr punc_listregex_remove_punctuations rrSentiText.__init__ s^$$$t{{7+,D "(@%#'#<#<#>  33D4L4LMrcURRSUR5nUR5nUVs1sHn[ U5S:dMUiM nn[ UR U5Vs0sHnSRU5US_M nn[ X R 5Vs0sHnSRU5US_M nnUnURU5 U$s snfs snfs snf)zD Returns mapping of form: { 'cat,': 'cat', ',cat': 'cat', } r) rsubrsplitlenrrjoinupdate)r no_punc_text words_onlywp punc_before punc_afterwords_punc_dicts r_words_plus_puncSentiText._words_plus_puncs4488TYYG !'') !+:As1vza :181TU1TArwwqz1Q4'1T U07 NN0ST0S1bggaj!A$&0S T%z* ;UTsC C 2C%*C*cURR5nUR5nUVs/sHn[U5S:dMUPM nn[ U5HupCX2;dM X#X'M U$s snf)z Removes leading and trailing puncutation Leaves contractions and most emoticons Does not preserve punc-plus-letter emoticons (e.g. :D) r)rrrr enumerate)rwesrweis rrSentiText._words_and_emoticons+sl iioo//1/Cb3r7Q;rC/s^EA$(,$ 0s A1A1cSnSnUHnUR5(dMUS- nM! [U5U- nSUs=:a[U5:aO U$SnU$)z Check whether just some words in the input are ALL CAPS :param list words: The words to inspect :returns: `True` if some but not all items in `words` are ALL CAPS FrrT)rr)rwords is_different allcap_wordsrcap_differentials rrSentiText.allcap_differential9sc  D||~~! u: 4  ,#e* , Lr)rrrrrN) rrrrrrrrrrrrrrrs N( rrcl\rSrSrSrSSjrSrSrSrSr Sr S r S r S r S rS rSrSrSrg)SentimentIntensityAnalyzeriKz0 Give a sentiment intensity score to sentences. c[RRU5UlUR 5Ul[ 5Ulgr)nltkdataload lexicon_file make_lex_dictlexiconr constants)rrs rr#SentimentIntensityAnalyzer.__init__Ps3!IINN<8))+ ')rc0nURRS5H4nUR5RS5SSup4[U5X'M6 U$)z& Convert lexicon file to a dictionary   rr)rrstripfloat)rlex_dictlinermeasures rr(SentimentIntensityAnalyzer.make_lex_dictXsU%%++D1D"jjl006q;OT"7^HN2rc[XRRURR5n/nURnUHnSnUR U5nU[ U5S- :a-UR5S:XaXGS-R5S:Xd(UR5URR;aURU5 MURXbXWU5nM URXC5nURX15$)a Return a float for sentiment strength based on the input text. Positive values are positive valence, negative value are negative valence. :note: Hashtags are not taken into consideration (e.g. #BAD is neutral). If you are interested in processing the text in the hashtags too, then we recommend preprocessing your data to remove the #, after which the hashtag text may be matched as if it was a normal word in the sentence. rrkindof) rrrrrindexrrrappendsentiment_valence _but_check score_valence)rr sentitext sentimentsritemrrs rpolarity_scores*SentimentIntensityAnalyzer.polarity_scoresbs ..**DNN,S,S  ';;'DG#))$/AC+,q00JJLF*'A.446$>!U(a7US:aXR R - nOXR R -n[SS5Hn XI:dM XtU S-- R5UR;dM2UR RXtU S-- X5n U S:Xa U S:waU S-n U S:Xa U S:waU S-n X-nURXX5nU S:XdMURXU5nM URXU5nURU5 U$)Nrrrgffffff?rg?) rrrrrrrranger _never_check _idioms_check _least_checkr) rrrrrrrritem_lowercasestart_iss rr,SentimentIntensityAnalyzer.sentiment_valencesd++ ';; \\ )ll>2G||~~+Q;~~444G~~444G A;K+1,=>DDF<<( 55+1,=>A!|QH!|QG%kG"//gG!|"&"4"4WST"U+'<''aHG'"rcUS:aX#S- R5UR;aeX#S- R5S:XaLX#S- R5S:wa1X#S- R5S:waXRR-nU$US:aTX#S- R5UR;a1X#S- R5S:XaXRR-nU$)Nrrrrrpr)rrrr)rrrrs rr 'SentimentIntensityAnalyzer._least_checks E#E*002$,,F#E*002g=$E*002d:'A.446&@!NN$;$;; E#E*002$,,F#E*002g= 7 77Grc UVs/sHo3R5PM nnS1[U5-nU(aSUR[[ U555n[ U5H!upgXe:a US-X&'MXe:dMUS-X&'M# U$s snf)Nbutg?r)rsetrnextiterr)rrrw_er(bisidx sentiments rr%SentimentIntensityAnalyzer._but_checks6IJ6Isyy{6IJg/00 $**4S ?;B#,Z#89'03J$Y'03J$ $9 KsB cX#S- SX#3nSRX#S- X#S- X#5nX#S- SX#S- 3nSRX#S- X#S- X#S- 5nSRX#S- X#S- 5nXEXgU/n U H7n XRR;dMURRU n O [U5S- U:a@X#SX#S-3n XRR;aURRU n[U5S- US-:aQSRX#X#S-X#S-5n XRR;aURRU nXRR;dX`RR;aXRR -nU$)Nr z{} {} {}rrz{} {})formatrrrrr) rrrronezero twoonezerotwoone threetwoonethreetwo sequencesseqzeroone zeroonetwos rr(SentimentIntensityAnalyzer._idioms_checks(Q/02E2H1IJ&& A & A &  " (A./q1DU1K0LM '' A & A & A & >> A &(;E(B &xH Cnn888..<URRX$S- /5(aXRR-nUS:XahX$S- S:XaX$S- S:Xd X$S- S:XaUS-nOAURRX$US-- /5(aXRR-nUS:XaX$S- S:XaX$S- S:Xd!X$S- S:XdX$S- S:Xd X$S- S:XaUS -nU$URRX$US-- /5(aXRR-nU$) Nrrrrrgthisrrg?)rrr)rrrr"rs rr'SentimentIntensityAnalyzer._never_checksS a<~~%%':q5'A&BCC!NN$;$;; a<"q5)W4#E*d2&1u-7!C-'')<'A+=N)O(PQQ!NN$;$;; a<#E*g5'A.$6*q51V;(A.$6*q51V;"D.'')<'A+=N)O(PQQ!NN$;$;;rcRURU5nURU5nX4-nU$r) _amplify_ep _amplify_qm)rsum_sr ep_amplifier qm_amplifierpunct_emph_amplifiers r_punctuation_emphasis0SentimentIntensityAnalyzer._punctuation_emphasis$s1''- ''- +:##rcBURS5nUS:aSnUS-nU$)Nrg㥛 ?count)rrep_countrEs rrB&SentimentIntensityAnalyzer._amplify_ep+s-::c? a<H %' rcVURS5nSnUS:aUS::aUS-nU$SnU$)Nrrrrg ףp= ?gQ?rL)rrqm_countrFs rrC&SentimentIntensityAnalyzer._amplify_qm5sB::c? a<1} ($  $ rcSnSnSnUH>nUS:aU[U5S-- nUS:aU[U5S- - nUS:XdM9US- nM@ X#U4$)Nrrr)r )rrpos_sumneg_sum neu_countsentiment_scores r_sift_sentiment_scores1SentimentIntensityAnalyzer._sift_sentiment_scoresBs{ )O"/*Q."/*Q.!#Q  ***rcU(Ga [[U55nURX25nUS:aX4- nO US:aX4-nURR U5nUR U5upgnU[ R"U5:aXd- nOU[ R"U5:aXt-nU[ R"U5-U-n [ R"Xi- 5n [ R"Xy- 5n [ R"X- 5n OSnSn Sn Sn [U S5[U S5[U S5[US5S.n U $)NrrrrK)negneuposcompound) r sumrHrrrXrfabsround)rrrrDrGr^rTrUrVtotalr]r[r\sentiment_dicts rr(SentimentIntensityAnalyzer.score_valenceTs2 #j/*E#'#=#=e#J qy--~~//6H*.*E*Ej*Q 'Gi7++/499W--/dii009rhsD  ccLAAHnnr