umbra/Irregular verbs.csv

+Infinitief,verleden tijd enkelvoud,verleden tijd meervoud,voltooid deelwoord
+bakken,bakte,bakten,gebakken
+bannen,bande,banden,gebannen
+barsten,barstte,barstten,gebarsten
+bederven,bedierf,bedierven,bedorven
+bedriegen,bedroog,bedrogen,bedrogen
+beginnen,begon,begonnen,begonnen
+begrijpen,begreep,begrepen,begrepen
+behangen,behing,behingen,behangen
+bergen,borg,borgen,geborgen
+bevallen,beviel,bevielen,bevallen
+bevelen,beval,bevalen,bevolen
+bewegen,bewoog,bewogen,bewogen
+bezoeken,bezocht,bezochten,bezocht
+bezwijken,bezweek,bezweken,bezweken
+bidden,bad,baden,gebeden
+bieden,bood,boden,geboden
+bijten,beet,beten,gebeten
+binden,bond,bonden,gebonden
+blazen,blies,bliezen,geblazen
+blijken,bleek,bleken,gebleken
+blijven,bleef,bleven,gebleven
+blinken,blonk,blonken,geblonken
+braden,braadde,braadden,gebraden
+breken,brak,braken,gebroken
+brengen,bracht,brachten,gebracht
+brouwen,brouwde,brouwden,gebrouwen
+buigen,boog,bogen,gebogen
+denken,dacht,dachten,gedacht
+dingen naar,dong naar,dongen naar,gedongen naar
+doen,deed,deden,gedaan
+dragen,droeg,droegen,gedragen
+drijven,dreef,dreven,gedreven
+dringen,drong,drongen,gedrongen
+drinken,dronk,dronken,gedronken
+druipen,droop,dropen,gedropen
+duiken,dook,doken,gedoken
+dwingen,dwong,dwongen,gedwongen
+ervaren,ervoer,ervoeren,ervaren
+eten,at,aten,gegeten
+fluiten,floot,floten,gefloten
+gaan,ging,gingen,gegaan
+gelden,gold,golden,gegolden
+genezen,genas,genazen,genezen
+genieten,genoot,genoten,genoten
+geven,gaf,gaven,gegeven
+gieten,goot,goten,gegoten
+glijden,gleed,gleden,gegleden
+glimmen,glom,glommen,geglommen
+graven,groef,groeven,gegraven
+grijpen,greep,grepen,gegrepen
+hangen,hing,hingen,gehangen
+hebben,had,hadden,gehad
+heffen,hief,hieven,geheven
+helpen,hielp,hielpen,geholpen
+heten,heette,heetten,geheten
+hijsen,hees,hesen,gehesen
+hoeven,hoefde,hoefden,gehoeven
+houden,hield,hielden,gehouden
+houwen,hieuw,hieuwen,gehouwen
+jagen,joeg,joegen,gejaagd
+jagen,jaagde,jaagden,gejaagd
+kiezen,koos,kozen,gekozen
+kijken,keek,keken,gekeken
+klimmen,klom,klommen,geklommen
+klinken,klonk,klonken,geklonken
+kluiven,kloof,kloven,gekloven
+knijpen,kneep,knepen,geknepen
+komen,kwam,kwamen,gekomen
+kopen,kocht,kochten,gekocht
+krijgen,kreeg,kregen,gekregen
+krimpen,kromp,krompen,gekrompen
+kruipen,kroop,kropen,gekropen
+kunnen,kon,konden,gekund
+lachen,lachte,lachten,gelachen
+laden,laadde,laadden,geladen
+laten,liet,lieten,gelaten
+lezen,las,lazen,gelezen
+liegen,loog,logen,gelogen
+liggen,lag,lagen,gelegen
+lijden,leed,leden,geleden
+lijken,leek,leken,geleken
+lopen,liep,liepen,gelopen
+malen,maalde,maalden,gemalen
+melken,molk,molken,gemolken
+melken,melkte,melkten,gemolken
+meten,mat,maten,gemeten
+mijden,meed,meden,gemeden
+moeten,moest,moesten,gemoeten
+mogen,mocht,mochten,gemogen
+neerzijgen,zeeg neer,zegen neer,neergezegen
+nemen,nam,namen,genomen
+nijgen,neeg,negen,genegen
+ontbijten,ontbeet,ontbeten,ontbeten
+ontginnen,ontgon,ontgonnen,ontgonnen
+ontluiken,ontlook,ontloken,ontloken
+ontwerpen,ontwierp,ontwierpen,ontworpen
+overlijden,overleed,overleden,overleden
+plegen,placht,plachten,gepleegd
+pluizen,ploos,plozen,geplozen
+prijzen,prees,prezen,geprezen
+raden,raadde,raadden,geraden
+raden,ried,rieden,geraden
+rijden,reed,reden,gereden
+rijgen,reeg,regen,geregen
+rijten,reet,reten,gereten
+rijzen,rees,rezen,gerezen
+roepen,riep,riepen,geroepen
+ruiken,rook,roken,geroken
+scheiden,scheidde,scheidden,gescheiden
+schelden,schold,scholden,gescholden
+schenden,schond,schonden,geschonden
+schenken,schonk,schonken,geschonken
+scheppen,shiep,schiepen,geschapen
+scheppen,schiep,schiepen,geschapen
+scheren,scheerde,scheerden,geschoren
+scheren,schoor,schoren,geschoren
+schieten,schoot,schoten,geschoten
+schijnen,scheen,schenen,geschenen
+schijten,scheet,scheten,gescheten
+schrijden,schreed,schreden,geschreden
+schrijven,schreef,schreven,geschreven
+schrikken,schrok,schrokken,geschrokken
+schuilen,school,scholen,gescholen
+schuilen,schuilde,schuilden,geschuild
+schuiven,schoof,schoven,geschoven
+slaan,sloeg,sloegen,geslagen
+slapen,sliep,sliepen,geslapen
+slijpen,sleep,slepen,geslepen
+slijten,sleet,sleten,gesleten
+slinken,slonk,slonken,geslonken
+sluipen,sloop,slopen,geslopen
+sluiten,sloot,sloten,gesloten
+smelten,smolt,smolten,gesmolten
+smijten,smeet,smeten,gesmeten
+snijden,sneed,sneden,gesneden
+snuiten,snoot,snoten,gesnoten
+snuiven,snoof,snoven,gesnoven
+spannen,spande,spanden,gespannen
+spijten,speet,speten,gespeten
+spinnen,spon,sponnen,gesponnen
+splijten,spleet,spleten,gespleten
+spreken,sprak,spraken,gesproken
+springen,sprong,sprongen,gesprongen
+spruiten,sproot,sproten,gesproten
+spuiten,spoot,spoten,gespoten
+staan,stond,stonden,gestaan
+steken,stak,staken,gestoken
+stelen,stal,stalen,gestolen
+sterven,stierf,stierven,gestorven
+stijgen,steeg,stegen,gestegen
+stijven,steef,steven,gesteven
+stinken,stonk,stonken,gestonken
+stoten,stootte,stootten,gestoten
+strijden,streed,streden,gestreden
+strijken,streek,streken,gestreken
+stuiven,stoof,stoven,gestoven
+treden,trad,traden,getreden
+treffen,trof,troffen,getroffen
+trekken,trok,trokken,getrokken
+vallen,viel,vielen,gevallen
+vangen,ving,vingen,gevangen
+varen,voer,voeren,gevaren
+vechten,vocht,vochten,gevochten
+verbieden,verbood,verboden,verboden
+verbinden,verbond,verbonden,verbonden
+verdrieten,verdroot,verdroten,verdroten
+verdrinken,verdronk,verdronken,verdronken
+verdwijnen,verdween,verdwenen,verdwenen
+vergelijken,vergeleek,vergeleken,vergeleken
+vergeten,vergat,vergaten,vergeten
+verlaten,verliet,verlieten,verlaten
+verliezen,verloor,verloren,verloren
+vermijden,vermeed,vermeden,vermeden
+verraden,verried,verrieden,verraden
+verraden,verraadde,verraadden,verraden
+verslinden,verslond,verslonden,verslonden
+vertrekken,vertrok,vertrokken,vertrokken
+verwijten,verweet,verweten,verweten
+verzinnen,verzon,verzonnen,verzonnen
+verzwelgen,verzwolg,verzwolgen,verzwolgen
+vinden,vond,vonden,gevonden
+vlechten,vlocht,vlochten,gevlochten
+vliegen,vloog,vlogen,gevlogen
+vouwen,vouwde,vouwden,gevouwen
+vragen,vroeg,vroegen,gevraagd
+vreten,vrat,vraten,gevreten
+vriezen,vroor,vroren,gevroren
+vriezen,vroor,-,gevroren
+vrijen,vree,vreeën,gevreeën
+vrijen,vrijde,vrijden,gevrijd
+wassen,waste,wasten,gewassen
+wassen,wies,wiesen,gewassen
+wegen,woog,wogen,gewogen
+werpen,wierp,wierpen,geworpen
+werven,wierf,wierven,geworven
+weten,wist,wisten,geweten
+weven,weefde,weefden,geweven
+wijken,week,weken,geweken
+wijten,weet,weten,geweten
+wijzen,wees,wezen,gewezen
+willen,wilde,wilden,gewild
+willen,wou,wouden,gewild
+winden,wond,wonden,gewonden
+winnen,won,wonnen,gewonnen
+worden,werd,werden,geworden
+wreken,wrak,wraken,gewroken
+wrijven,wreef,wreven,gewreven
+wringen,wrong,wrongen,gewrongen
+zeggen,zei,zeiden,gezegd
+zenden,zond,zonden,gezonden
+zich kwijten van,kweet zich van,kweten zich van,zich gekweten van
+zien,zag,zagen,gezien
+zijn,was,waren,geweest
+zingen,zong,zongen,gezongen
+zinken,zonk,zonken,gezonken
+zinnen,zon,zonnen,gezonnen
+zitten,zat,zaten,gezeten
+zoeken,zocht,zochten,gezocht
+zouten,zoutte,zoutten,gezouten
+zuigen,zoog,zogen,gezogen
+zuipen,zoop,zopen,gezopen
+zullen,zou,zouden,-
+zwelgen,zwolg,zwolgen,gezwolgen
+zwellen,zwol,zwollen,gezwollen
+zwemmen,zwom,zwommen,gezwommen
+zweren,zwoer,zwoeren,gezworen
+zwerven,zwierf,zwierven,gezworven
+zwijgen,zweeg,zwegen,gezwegen

umbra/controller.py

 from filereader import FileReader, CSVReader, TxtReader
 from filereader import FileWriter, CSVWriter, TxtWriter
 from utils import Utils as ut
-import tkinter.filedialog as fd
 import os
 
 
@@ -57,12 +56,14 @@ class Controller:
         Select a folder from which multiple files can be read.
         """
         folder_path = self._view.dir_dialog()
-        file_names = os.listdir(folder_path)
-        file_paths = [folder_path + "/" + x for x in file_names]
-        self._set_path(file_paths, type)
-        self._view.update_files(file_paths, type)
-        if file_paths:
-            self._view.button_status("select {}".format(type), "disabled")
+        if folder_path:
+            file_names = os.listdir(folder_path)
+            file_paths = [folder_path + "/" + x for x in file_names \
+                            if ".csv" in x]
+            self._set_path(file_paths, type)
+            self._view.update_files(file_paths, type)
+            if file_paths:
+                self._view.button_status("select {}".format(type), "disabled")
         # TODO: Call 'buttons' by functionality, not object (type)
     def _set_path(self, paths, type):
         if type == 'source':

umbra/dutchwordnet_test.py

-from OpenDutchWordnet.wn_grid_parser import Wn_grid_parser
-from semantic_checker import SemanticChecker
-
-instance = Wn_grid_parser("OpenDutchWordnet/resources/odwn/odwn_orbn_gwg-LMF_1.3.xml.gz")
-instance.load_synonyms_dicts()
-sc = SemanticChecker(instance)
-#A test case for when the second word is a hypernym of the first
-print(sc.semantically_related('hond','carnivoor'))
-#A test case for when the first word is a hypernym of the second
-print(sc.semantically_related('carnivoor','hond'))
-#A test case for when both words are known yet not related.
-print(sc.semantically_related('water','hond'))
-#A test case for when both words are known and considered synonyms
-print(sc.semantically_related('huis','woning'))
-#A test case for when the second word is unknown
-print(sc.semantically_related('hond','atestword'))
-#A test case for when both words are unknown
-print(sc.semantically_related('testfirstword','sometestword'))
-#A test case for when the first word is unknown
-print(sc.semantically_related('xxxxxxx','carnivoor'))
-#A test case for when the two words share a hypernym
-print(sc.semantically_related('vork','mes'))
-
-#Other test cases to check for other words:
-print('And now for the other test cases:')
-print(sc.semantically_related('het','de'))
-print(sc.semantically_related('een','ander'))
-print(sc.semantically_related('woord','zin'))
-print(sc.semantically_related('naam','persoon'))
-print(sc.semantically_related('MSDT','project'))
-print(sc.semantically_related('dog','pet'))
-
-
-
-
-

umbra/form_checker.py

+import pandas as pd
+from bisect import bisect_left
+from utils import Utils as ut
+
+class FormChecker:
+
+    def __init__(self):
+        """
+        Constructor
+        """
+        self.initialize_frames()
+        self._prefixes = ['','ge','be','ver','on','ont']
+        self._affixes = ['','en','t','te','ten','de','den','s',"'s"]
+
+    def initialize_frames(self):
+        """
+        Help the constructor by initializing the frames for irregular verbs.
+        """
+        self._irr_verbs_frame = pd.read_csv(ut.get_path('Irregular verbs.csv'))
+        self._irr_verbs_vte = self._irr_verbs_frame.copy()
+        self._irr_verbs_vtm = self._irr_verbs_frame.copy()
+        self._irr_verbs_vdw = self._irr_verbs_frame.copy()
+        self._irr_verbs_frame = self._irr_verbs_frame.values
+        self._irr_verbs_vte = self._irr_verbs_vte.sort_values(
+            'verleden tijd enkelvoud').values
+        self._irr_verbs_vtm = self._irr_verbs_vtm.sort_values(
+            'verleden tijd meervoud').values
+        self._irr_verbs_vdw = self._irr_verbs_vdw.sort_values(
+            'voltooid deelwoord').values
+
+    def form_related(self, source_word, shadow_word):
+        """
+        Check whether the arguments can be considered to be form-related.
+
+        Args:
+            source_word: lemma of type string, corresponding to a source word
+            shadow_word: lemma of type string, corresponding to a shadow word
+
+        Returns:
+            A boolean value indicating whether the arguments are form-related
+        """
+        return self.fix_related(source_word,shadow_word) or \
+               self.irr_verb_related(source_word,shadow_word)
+
+    def irr_verb_related(self, source_word, shadow_word):
+        """
+        Check whether the arguments are different versions of the same
+        irregular verb.
+
+        Args:
+            source_word: lemma of type string, corresponding to a source word
+            shadow_word: lemma of type string, corresponding to a shadow word
+
+        Returns:
+            A boolean value indicating whether the arguments are different
+            versions of the same irregular verb.
+        """
+        if self.irr_verb_helper(self._irr_verbs_frame, 0,
+                                source_word,shadow_word):
+            return True
+        elif self.irr_verb_helper(self._irr_verbs_vte,1
+                                  , source_word,shadow_word):
+            return True
+        elif self.irr_verb_helper(self._irr_verbs_vtm,2
+                                  , source_word,shadow_word):
+            return True
+        elif self.irr_verb_helper(self._irr_verbs_vdw,3
+                                  , source_word,shadow_word):
+            return True
+        return False
+
+    def irr_verb_helper(self, frame, key, source_word, shadow_word):
+        """
+        Search in one frame to see whether the shadow word and source word are
+        in the same row.
+
+        Args:
+            frame: A numpy array with String types
+            key: The column number in which to look
+            source_word: lemma of type string, corresponding to a source word
+            shadow_word: lemma of type string, corresponding to a shadow word
+
+        Returns:
+            A boolean value indicating whether the source word could be found
+            and the shadow word is in the same row as the source word.
+
+        """
+        i = bisect_left(frame[key], source_word)
+        if i != len(frame[key]) and shadow_word in frame[i]:
+            return True
+        return False
+
+    def reg_verb_related(self, source_word, shadow_word):
+        pass
+
+    def fix_related(self,source_word,shadow_word):
+        """
+        Check whether the arguments are related in terms of their prefixes and
+        affixes.
+
+        Args:
+            source_word: lemma of type string, corresponding to a source word
+            shadow_word: lemma of type string, corresponding to a shadow word
+
+        Returns:
+            A boolean value indicating whether the arguments are related in
+            terms of their prefixes and affixes.
+        """
+        return self.prefix_related(source_word,shadow_word) or \
+            self.affix_related(source_word,shadow_word)
+
+    def prefix_related(self, source_word, shadow_word):
+        """
+        Check whether the arguments are related in terms of their prefixes
+
+        Args:
+            source_word: lemma of type string, corresponding to a source word
+            shadow_word: lemma of type string, corresponding to a shadow word
+
+        Returns:
+            A boolean value indicating whether the arguments are related in
+            terms of their prefixes.
+        """
+        for pre in self._prefixes:
+            len_pre = len(pre)
+            if shadow_word[:len_pre] == pre:
+                shadow_rest = shadow_word[len_pre:]
+                for pre2 in self._prefixes:
+                    if pre2 + shadow_rest == source_word:
+                        return True
+        return False
+
+    def affix_related(self, source_word, shadow_word):
+        """
+        Check whether the arguments are related in terms of their prefixes and
+        postfixes.
+
+        Args:
+            source_word: lemma of type string, corresponding to a source word
+            shadow_word: lemma of type string, corresponding to a shadow word
+
+        Returns:
+            A boolean value indicating whether the arguments are related in
+            terms of their prefixes and postfixes.
+        """
+        for aff in self._affixes:
+            len_aff = len(aff)
+            if shadow_word[-len_aff:] == aff:
+                shadow_rest = shadow_word[:-len_aff]
+                for aff2 in self._affixes:
+                    if shadow_rest + aff2 == source_word:
+                        return True
+        return False
\ No newline at end of file

umbra/form_tester.py

+from form_checker import FormChecker
+import time
+
+former = FormChecker()
+print(former.irr_verb_related('bakken', 'bakten'))
+print(former.fix_related('hyena', "hyena's"))
+print(former.fix_related('hyena',"gehyena"))
+print(former.fix_related('gehyena','verhyena'))
+print(former.fix_related('hyenaten','hyenaden'))
+print(former.fix_related('gehyena','verhyenas'))

umbra/mistake_enum.py

@@ -16,3 +16,4 @@ class Mistake(Enum):
     SEMANTIC = "semantic"
     SKIPPED = "skipped"  # For source words that are not shadowed
     RANDOM = "random"  # For shadow words that do not reflect a source word
+    FORM = "form"   # For verbs that are shadowed in another form

umbra/model.py

@@ -50,7 +50,7 @@ class Model:
     def id(self):
         return self._id
 
-    @id.setter
+    @id.setter # TODO: DO we still use this? As far as I have seen, we do not. Also, change name of id here.
     def id(self, id):
         self._id = id
 
@@ -70,14 +70,6 @@ class Model:
     def data_source(self, data):
         self._data_source = data
 
-    @property
-    def multi_data_shadow(self):
-        return self._multi_data_shadow
-
-    @multi_data_shadow.setter
-    def multi_data_shadow(self, data):
-        self._multi_data_shadow = data
-
     @property
     def multi_data_source(self):
         return self._multi_data_source
@@ -86,18 +78,6 @@ class Model:
     def multi_data_source(self, data):
         self._multi_data_source = data
 
-    @property
-    def data_shadow(self):
-        return self._data_shadow
-
-    @data_shadow.setter
-    def data_shadow(self, data):
-        self._data_shadow = data
-
-    @property
-    def analysis_results(self):
-        return self._analysis_results
-
     def has_source(self):
         """Check whether self._data_source has a value."""
         return self._data_source is not None \
@@ -112,11 +92,13 @@ class Model:
         """Check whether self._analysis_results has a value."""
         return self._analysis_results or self._multi_results
 
-    def get_multi_data(self, type):
-        """
-        Get data according to type
+    def get_multi_data(self, data_type): #TODO: Why do we have this as well when you can also just get multi_data_source itself?
+        """Get data according to type
+
+        Args:
+            data_type: the type of data to get (either source or shadow)
         """
-        if type == "source":
+        if data_type == "source":
             return self._multi_data_source
         else:
             return self._multi_data_shadow

umbra/needleman_wunsch.py

@@ -9,21 +9,24 @@ class NeedlemanWunsch(AlignmentStrategy):
     """Class that aligns the words of a shadow and source file according
     to the Needleman-Wunsch algorithm"""
 
-    def __init__(self, seman_checker):
+    def __init__(self, seman_checker, form_checker):
         super().__init__()
         self._match = 4
         self._mismatch = -2
         self._gap_sc = -1
         self._seman_match = 2
         self._repetition = 0
+        self._form_match = 2
         self._pointers = ['diag', 'up', 'left']
         self._source = None
         self._shadow = None
         self._matrix = None
         self._seman_checker = seman_checker
+        self._form_checker = form_checker
 
     def alignment_options(self, match=None, mismatch=None,
-                gap_sc=None, seman_match=None, repetition=None):
+                          gap_sc=None, seman_match=None, repetition=None,
+                          form_match=None):
         """ Set the scores that are allocated whilst aligning. Can be changed
             one at a time or more at once.
 
@@ -46,6 +49,8 @@ class NeedlemanWunsch(AlignmentStrategy):
             self._seman_match = seman_match
         if repetition:
             self._repetition = repetition
+        if form_match:
+            self._form_match = form_match
 
     def align(self, source, shadow):
         """ This is the main function of finding alignments.
@@ -66,7 +71,6 @@ class NeedlemanWunsch(AlignmentStrategy):
         self._fill_matrix()
         alignment_source, alignment_shadow = self._traceback()
         return self._source, self._shadow
-        # return alignment_source, alignment_shadow
 
     def _initialize_matrix(self):
         """ Initialize matrix according to the Needleman-Wunsch algorithm.
@@ -95,6 +99,9 @@ class NeedlemanWunsch(AlignmentStrategy):
             for j in range(1, n+1):
                 if self._source[j - 1] == self._shadow[i - 1]:
                     value = self._match
+                elif self._form_checker.form_related(self._source[j-1].word,
+                                                     self._shadow[i-1].word):
+                    value = self._form_match
                 elif self._seman_checker.semantically_related(
                         self._source[j-1].word, self._shadow[i-1].word):
                     value = self._seman_match
@@ -155,14 +162,16 @@ class NeedlemanWunsch(AlignmentStrategy):
         return Sentence(alignment_source), Sentence(alignment_shadow)
 
     def _check_repetition(self, index):
-        """ Check whether a word is repeated or not
+        """ Check whether a word is repeated or not.
 
         Args:
-            index: the index of the word to check if it is a repeat
+            index: The index of the word to check
         """
+        # TODO: check for form mistakes first
         word = self._shadow[index]
-        return index > 0 and word == self._shadow[index-1] or \
-            index < len(self._shadow) - 1 and word == self._shadow[index + 1]
+        return index > 0 and self._shadow[index - 1].word.endswith(word.word) \
+            or index < len(self._shadow) - 1 and \
+            self._shadow[index + 1].word.startswith(word.word)
 
     def _equals_checker(self, source, shadow):
         """ Check whether two words are syntactically or semantically equal
@@ -176,8 +185,13 @@ class NeedlemanWunsch(AlignmentStrategy):
             source.shadowed = True
             shadow.correct = True
             shadow.source = source
-        elif self._seman_checker.semantically_related(source.word,shadow.word):
+        elif self._seman_checker.semantically_related(source.word,
+                                                      shadow.word):
             source.shadowed = True
             source.mistake = Mistake.SEMANTIC
             shadow.mistake = Mistake.SEMANTIC
+        elif self._form_checker.form_related(source.word, shadow.word):
+            source.shadowed = True
+            source.mistake = Mistake.FORM
+            shadow.mistake = Mistake.FORM
         return source, shadow

umbra/statistics.py

@@ -3,6 +3,7 @@ from needleman_wunsch import NeedlemanWunsch
 from anchor_algorithm import AnchorAlgorithm
 from mistake_finder import MistakeFinder
 from semantic_checker import SemanticChecker
+from form_checker import FormChecker
 from utils import Utils as ut
 import time
 
@@ -18,10 +19,10 @@ class Statistics:
         # currently the case, so the two operate seperately in this class.
         # _parser and _seman_checker should be moved to MistakeFinder when
         # possible
-        self.path = "OpenDutchWordnet/resources/odwn/odwn_orbn_gwg-LMF_1.3.xml.gz"
-        # "./umbra/OpenDutchWordnet/resources/odwn/odwn_orbn_gwg-LMF_1.3.xml.gz")  # Ubuntu path, temporary 'fix'
+        self.path = ut.get_path("OpenDutchWordnet/resources/odwn/odwn_orbn_gwg-LMF_1.3.xml.gz")
 
         self._seman_checker = SemanticChecker()
+        self._form_checker = FormChecker()
         self._mistake_finder = MistakeFinder(self._seman_checker)
 
     @property
@@ -50,7 +51,8 @@ class Statistics:
         print('\n Needleman-Wunsch')
         results = 'Needleman-Wunsch'
         discrete_start_time = time.time()
-        self._strategy = NeedlemanWunsch(self._seman_checker)
+        self._strategy = NeedlemanWunsch(self._seman_checker,
+                                         self._form_checker)
         source_align, shadow_align = self._strategy.align(source, shadow)
         for s_word in shadow_align:
             if s_word.has_source():

umbra/umbra.py

@@ -7,8 +7,8 @@ class Umbra:
     """Initialise seperate components of program."""
 
     def __init__(self):
-        self._model = Model()
         self._view = View()
+        self._model = Model()
         self._controller = Controller(self._view, self._model)
 
     def main(self):

umbra/utils.py

@@ -87,3 +87,4 @@ class Utils:
                         cnt += 1
                 key += "({})".format(str(cnt))
                 dict[key] = value
+            pass

umbra/view.py

@@ -4,18 +4,28 @@ from tkinter import ttk, W, filedialog  # tk.ttk yields an AttributeError
 from functools import partial
 from utils import Utils as ut
 import tkinter.filedialog as fd
+from PIL import ImageTk, Image
+
 
 class View:
     """Handle GUI and I/O elements of the program."""
 
     def __init__(self):
         self._window = tk.Tk()
+
+        # Center window
+        self._center_window()
+
+        # Draw splash
+        self._splash = SplashView(self._window)
+
+        # Create rest of GUI
         self._window.wm_title("Umbra")
         self._frame = tk.Frame(self._window).grid(column=0, row=0)
         self._elements = {}  # Dict to avoid large number of attributes
         self._selected_source = "No file selected"
         self._selected_shadow = "No file selected"
-        # self._options = Options(self)
+        #self._options = Options(self)
         # Options are not redundant quite yet
         self._actionlistener = None
 
@@ -44,6 +54,9 @@ class View:
         self._create_label('message', self._frame, "", 5, 1)
         self._spit = False
 
+        # Window Icon
+        self._window.iconbitmap("./resources/logo.ico")
+
     @property
     def spit(self):
         return self._spit
@@ -54,6 +67,8 @@ class View:
 
     def display(self):
         """Start main loop, displaying GUI elements."""
+        #self._splash.terminate()
+        self._window.deiconify()
         self._window.mainloop()
 
     def select_files(self, file_type):
@@ -98,12 +113,14 @@ class View:
         if paths == []:
             filebox.set("No file selected")
             filebox['values'] = []
+            filebox['state'] = "disabled"
         else:
             filenames = []
             for path in paths:
                 filenames.append(path.title().split("/")[-1])
             filebox['values'] = filenames
             filebox.set(filenames[-1])
+            filebox['state'] = "read_only"
 
         # Easier to just update both than to do a type check AGAIN
         self._selected_source = self._elements['file source'].get()
@@ -172,6 +189,16 @@ class View:
 
     ### Element creation & placement  # TODO: Unify arguments (especially positional)
 
+    def _center_window(self):
+        self._window.update_idletasks()
+        self._pos_right = int(self._window.winfo_screenwidth() / 2
+                              - self._window.winfo_reqwidth())
+        self._pos_down = int(self._window.winfo_screenheight() / 2
+                             - self._window.winfo_reqheight())
+        self._window.geometry("+{}+{}".format(self._pos_right, self._pos_down))
+        self._window.update()
+        self._window.withdraw()
+
     def _menu_bar(self, key="menu"):  # we only need the one, so...
         pass
 
@@ -198,20 +225,21 @@ class View:
                                       type,
                                       2, column)
         self._create_button('select {}'.format(type),
-                           frame,
-                           'Select {} file'.format(type),
-                           5, column)
+                            frame,
+                            'Add {} file'.format(type),
+                            5, column)
         self._create_button('delete {}'.format(type),
-                           frame,
-                           'Delete {}'.format(type),
-                           7, column)
+                            frame,
+                            'Remove {}'.format(type),
+                            7, column)
         self._create_button('rm_all {}'.format(type),
-                           frame,
-                           'Delete all {} files'.format(type),
-                           8, column)
-        self._create_button('select_folder {}'.format(type), frame,
-                                'Select {} folder'.format(type),
-                                6, column)
+                            frame,
+                            'Remove all {} files'.format(type),
+                            8, column)
+        self._create_button('select_folder {}'.format(type),
+                            frame,
+                            'Add {} folder'.format(type),
+                            6, column)
 
     def _create_button(self, key, frame, text, row, column):
         """Create button and place it in the given frame.
@@ -247,7 +275,7 @@ class View:
             ValueError: Key already exists
         """
         filebox = ttk.Combobox(frame,
-                               state="readonly",
+                               state="disabled",
                                width=20)
         filebox.set("No file selected")
 
@@ -349,3 +377,32 @@ class Options(View):
         action_function = partial(self._view._perform_action, key=command)
         tab.add_command(label=label)
         tab.bind('<Button>', action_function)
+
+
+class SplashView(tk.Toplevel):
+    """A splash screen while the rest of the application is loading. """
+
+    def __init__(self, parent):
+        tk.Toplevel.__init__(self, parent)
+        self.grab_set()
+        self.title("Umbra")
+        self.iconbitmap(ut.get_path("resources/logo.ico"))
+        im_path = Image.open(ut.get_path("resources/splash.png"))
+        ph = ImageTk.PhotoImage(im_path)
+        load_img = tk.Label(self, image=ph)
+        load_img.grid(row=0, column=0)
+
+        # Center window and remove window border
+        # TODO: Currently this block is mostly a duplicate of _center_window
+        self.update_idletasks()
+        self._pos_right = int(self.winfo_screenwidth() / 2
+                              - self.winfo_reqwidth() + 150)
+        self._pos_down = int(self.winfo_screenheight() / 2
+                             - self.winfo_reqheight())
+        self.geometry("+{}+{}".format(self._pos_right, self._pos_down))
+        self.overrideredirect(True)  # TODO: This currently breaks on Ubuntu
+
+        self.update()
+
+    def terminate(self):
+        self.destroy()