Merge branch 'master' of github.com:inpho/topic-explorer

ipynb/Topic-Explorer-Tutorial.py2.ipynb

@@ -304,7 +304,7 @@
     "The above code shows the topic-word distributions and allows us to estimate the quality of our topics.\n",
     "\n",
     "#### `v.labels`\n",
-    "The property `v.labels` (without parentheses) returns a list of all documents in a corpus, and is useful for processing each document generically, wihtout having to look up the identifiers on the file system.\n",
+    "The property `v.labels` (without parentheses) returns a list of all documents in a corpus, and is useful for processing each document generically, without having to look up the identifiers on the file system.\n",
     "\n",
     "Below, we print the first 3 document labels:"
    ]
@@ -386,7 +386,7 @@
     "#### Alternative distance measures\n",
     "By default, the Topic Explorer uses the Jensen-Shannon Distance to calculate the distance between documents. The Jensen-Shannon Distance (JSD) is a symmetric measure based on information theory that characterizes the difference between two probability distributions.\n",
     "\n",
-    "However, several alternate methods are built into the `vsm.spatial` module. These include the Kullbeck-Liebler Divergence, which is an asymmetric component of the JSD and is used in [Murdock et al. (in review)](http://arxiv.org/abs/1509.07175) to characterize the cognitive surprise of a new text, given previous texts.\n",
+    "However, several alternate methods are built into the `vsm.spatial` module. These include the Kullback-Leibler Divergence, which is an asymmetric component of the JSD and is used in [Murdock et al. (in review)](http://arxiv.org/abs/1509.07175) to characterize the cognitive surprise of a new text, given previous texts.\n",
     "\n",
     "Rather than using the JSD and assuming symmetric divergence between items, we assume that the second document is encountered after the first, effectively measuring text-to-text divergence."
    ]

ipynb/Topic-Explorer-Tutorial.py3.ipynb

@@ -344,7 +344,7 @@
     "The above code shows the topic-word distributions and allows us to estimate the quality of our topics.\n",
     "\n",
     "#### `v.labels`\n",
-    "The property `v.labels` (without parentheses) returns a list of all documents in a corpus, and is useful for processing each document generically, wihtout having to look up the identifiers on the file system.\n",
+    "The property `v.labels` (without parentheses) returns a list of all documents in a corpus, and is useful for processing each document generically, without having to look up the identifiers on the file system.\n",
     "\n",
     "Below, we print the first 3 document labels:"
    ]
@@ -426,7 +426,7 @@
     "#### Alternative distance measures\n",
     "By default, the Topic Explorer uses the Jensen-Shannon Distance to calculate the distance between documents. The Jensen-Shannon Distance (JSD) is a symmetric measure based on information theory that characterizes the difference between two probability distributions.\n",
     "\n",
-    "However, several alternate methods are built into the `vsm.spatial` module. These include the Kullbeck-Liebler Divergence, which is an asymmetric component of the JSD and is used in [Murdock et al. (2017)](http://arxiv.org/abs/1509.07175) to characterize the cognitive surprise of a new text, given previous texts.\n",
+    "However, several alternate methods are built into the `vsm.spatial` module. These include the Kullback-Leibler Divergence, which is an asymmetric component of the JSD and is used in [Murdock et al. (2017)](http://arxiv.org/abs/1509.07175) to characterize the cognitive surprise of a new text, given previous texts.\n",
     "\n",
     "Rather than using the JSD and assuming symmetric divergence between items, we assume that the second document is encountered after the first, effectively measuring text-to-text divergence."
    ]