Update curated.py

curated.py

@@ -529,7 +529,6 @@ filtering_process = Div(
         H3("Wikipedia"),
         P("Wikipedia is an encyclopedia form of high-quality text data used for language modeling. We have included filtered and deduplicated versions of complete Wikipedia data directly provided by the Wikipedia Foundation for more than 350 languages."),
         P(B("Download and Extraction: "), "The Wikimedia dataset was downloaded from the official snapshot on Huggingface: ", A("https://huggingface.co/datasets/wikimedia/wikipedia/tree/main", href="https://huggingface.co/datasets/wikimedia/wikipedia/tree/main"), ". The", D_code("huggingface dataset.to_json", language="python"), " function was used to convert the original parqet format to the jsonl format."),
-
         P(B("Filtering: "), "Manual inspection of the dataset demostrated high quality content. Only one filter was used to remove articles with few words. Based normal sentence constructs, the article was kept if it contained 10 or more words. Any article with fewer than 10 words was removed."),
         table_div_wikipedia,
         Details(
@@ -551,15 +550,13 @@ filtering_process = Div(
         Div(
         H3("ArXiv"),
         P("The ArXiv dataset is a vast collection of preprint research papers primarily in Mathematics, Computer Science, and Physics. Established in 1991, it offers high-quality text and mathematical knowledge, making it an invaluable resource for academic and scientific research. ArXiv papers are typically written in LaTeX, a popular typesetting system for these fields. We have extracted the information from latex and converted it into a text format."),
-        
         P(B("Download and Extraction: "),"All the data was downloaded in original latex format from Arxiv official S3 dump ", A("s3://arxic/src", href="s3://arxic/src"), ". We try to encode the downloaded data into utf-8 or guess encoding using chardet library. After that pandoc was used to extract information from the latex files and saved as markdown format",  D_code("pandoc -s {tex} -o out/{out_name}.md --wrap=none", language="python"), ". All markdowns were combined to create jsonl files."),
-        H4("Filtering"),
-        P("Multiple filters are used here after manually verifying output of all the filters as suggested by peS2o dataset (citation needed)"),
-        Ol(
-            Li("Language Filter: any language other than English are discarded"),
-            Li("Minimum Word Count Filter: less than 500 words (not inclusive) are discarded"),
-            Li("Unigram Log Probablity Filter Theshold: -20"),
-            Li("Note: the Frequency Filter was calculated but not applied. The most frequent word in the paper consists of alpha characters only, and it appears in less than 7.5% of the document. Words are obtained by splitting the text on whitespace."),
+        P(B(" Filters Applied: "), "multiple filters are used here after manually verifying output of all the filters as suggested by peS2o dataset (citation needed)"),
+        Ul(
+            Li("Language Filter: any language other than English are discarded", style = "margin-bottom: -3px"),
+            Li("Minimum Word Count Filter: less than 500 words (not inclusive) are discarded", style = "margin-bottom: -3px"),
+            Li("Unigram Log Probablity Filter Theshold: -20", style = "margin-bottom: -3px"),
+            Li("Note: the Frequency Filter was calculated but not applied. The most frequent word in the paper consists of alpha characters only, and it appears in less than 7.5% of the document. Words are obtained by splitting the text on whitespace.", style = "margin-bottom: -3px"),
         ),
         table_div_arx,
         Details(
@@ -581,18 +578,15 @@ filtering_process = Div(
         Div(
         H3("S2ORC"),
         P("The Semantic Scholar Open Research Corpus (S2ORC) is a comprehensive dataset designed for natural language processing (NLP) and text-mining research over scientific papers. It includes rich metadata, and abstract and full-text content for millions of academic papers across various disciplines. This dataset is further divided into two components, S2ORC abstract and S2ORC full text."),
-        H4("Download and Extraction"),
-        Ol(
-            Li("This was downloaded directly in zip format using S2ORC api key and a get() request: ", D_code("response = urllib.request.urlopen(url)", language = "python")),
-        ),
-        H4("Filtering"),
-        P("1. Multiple filters are used here after manually verifying output of all the filters as suggested by peS2o dataset"),
-        Ol(
-            Li("Title and Abstract Filter: must have title and abstract"),
-            Li("Language Filter: The paper must be in English. To determine the language of each document, we use the pycld3 library. We run pycld3 on the first 2000 characters of each paragraph in the paper. The language of the paper is the most common language of the paragraphs."),
-            Li("Word Count Filter: less than 500 words (not inclusive) are discarded"),
-            Li("Paragraph Count Filter: The paper must have at least 5 paragraphs after removing paragraphs with less than -20 average log world probability"),
-            Li("Frequency Filter: The most frequent word in the paper consists of alpha characters only, and it appears in less than 7.5% of the document. Words are obtained by splitting the text on whitespace."),
+        H4(""),
+        P(B("Download and Extraction: "),"S2ORC was downloaded directly in zip format using S2ORC api key and a get() request: ", D_code("response = urllib.request.urlopen(url)", language = "python")),
+        P(B("Filters Applied: "), "Multiple filters are used here after manually verifying output of all the filters as suggested by peS2o dataset"),
+        Ul(
+            Li("Title and Abstract Filter: must have title and abstract", style = "margin-bottom: -3px"),
+            Li("Language Filter: The paper must be in English. To determine the language of each document, we use the pycld3 library. We run pycld3 on the first 2000 characters of each paragraph in the paper. The language of the paper is the most common language of the paragraphs.", style = "margin-bottom: -3px"),
+            Li("Word Count Filter: less than 500 words (not inclusive) are discarded", style = "margin-bottom: -3px"),
+            Li("Paragraph Count Filter: The paper must have at least 5 paragraphs after removing paragraphs with less than -20 average log world probability", style = "margin-bottom: -3px"),
+            Li("Frequency Filter: The most frequent word in the paper consists of alpha characters only, and it appears in less than 7.5% of the document. Words are obtained by splitting the text on whitespace.", style = "margin-bottom: -3px"),
         ),
         table_div_s2o,
         Details(
@@ -614,18 +608,15 @@ filtering_process = Div(
         Div(
         H3("S2ORC Abstract"),
         P("The Semantic Scholar Open Research Corpus (S2ORC) is a comprehensive dataset designed for natural language processing (NLP) and text-mining research over scientific papers. It includes rich metadata, and abstract and full-text content for millions of academic papers across various disciplines. This dataset is further divided into two components, S2ORC abstract and S2ORC full text."),
-        H4("Download and Extraction"),
-        Ol(
-            Li("This was downloaded directly in zip format using S2ORC API key and a get() request: ", D_code("response = urllib.request.urlopen(url)", language = "python")),
-        ),    
-        H4("Filtering"),
-        P("Multiple filters are used here after manually verifying output of all the filters as suggested by peS2o dataset. The frequency filter was not used as suggested by peS2o because it was removing good samples as inspected manually"),
-        Ol(
-            Li("Title and Abstract Filter: must have title and abstract"),
-            Li("Majority Language Filter: abstract must be in English"),
-            Li("Minimum Word Count Filter: less than 20 (not inclusive) are discarded"),
-            Li("Unigram Log Probability Threshold: -20"),
-            Li("Note: Frequency Filter: The most frequent word in the paper consists of alpha characters only, and it appears in less than 7.5% of the document. Words are obtained by splitting the text on whitespace."),
+        P(B("Download and Extraction: "),"S2ORC was downloaded directly in zip format using S2ORC api key and a get() request: ", D_code("response = urllib.request.urlopen(url)", language = "python")),  
+        
+        P(B("Filters Applied: "), "multiple filters are used here after manually verifying output of all the filters as suggested by peS2o dataset. The frequency filter was not used as suggested by peS2o because it was removing good samples as inspected manually"),
+        Ul(
+            Li("Title and Abstract Filter: must have title and abstract", style = "margin-bottom: -3px"),
+            Li("Majority Language Filter: abstract must be in English", style = "margin-bottom: -3px"),
+            Li("Minimum Word Count Filter: less than 20 (not inclusive) are discarded", style = "margin-bottom: -3px"),
+            Li("Unigram Log Probability Threshold: -20", style = "margin-bottom: -3px"),
+            Li("Note: Frequency Filter: The most frequent word in the paper consists of alpha characters only, and it appears in less than 7.5% of the document. Words are obtained by splitting the text on whitespace.", style = "margin-bottom: -3px"),
         ),
         Details(
             Summary("S2ORC Abstract Filtering Examples "),
@@ -647,14 +638,13 @@ filtering_process = Div(
     Section(
         Div(
         H3("PubMed Central and PubMed Abstract"),
-            P(B("Download and Extraction: "), "All files were downloaded from", A("ttps://ftp.ncbi.nlm.nih.gov/pub/pmc/oa_package/",href="ttps://ftp.ncbi.nlm.nih.gov/pub/pmc/oa_package/"),". PubMed Central (PMC) files are downloaded in an xml.tar format. The tar files are opened and converted to markdown format using pandoc", D_code("pandoc -f jats {nxml} -o {pmcid}.md", language="bash"),". The markdown files are combined to create jsonl files. PubMed Abstract (PMA) files  were downloaded in xml. The BeautifulSoup library was used to extract the abstract, title, and PMID. All files were stored in jsonl format."),
-        H4("Filtering"),
-        P("1. Multiple filters are used here after manually verifying output of all the filters as suggested by peS2o dataset."),
-        Ol(
-            Li("Minimum Word Count Filter: less than 100 words (not inclusive) are discarded, less than 20 words for pubmed abstract"),
-            Li("Language Filter: any language other than English are discarded"),
-            Li("Frequency Filter: The most frequent word in the paper consists of alpha characters only, and it appears in less than 7.5% of the document. Words are obtained by splitting the text on whitespace. This filter is not used for pubmed abstract"),
-            Li("Unigram Log Probability Threshold: -20"),
+        P(B("Download and Extraction: "), "All files were downloaded from", A("ttps://ftp.ncbi.nlm.nih.gov/pub/pmc/oa_package/",href="ttps://ftp.ncbi.nlm.nih.gov/pub/pmc/oa_package/"),". PubMed Central (PMC) files are downloaded in an xml.tar format. The tar files are opened and converted to markdown format using pandoc", D_code("pandoc -f jats {nxml} -o {pmcid}.md", language="bash"),". The markdown files are combined to create jsonl files. PubMed Abstract (PMA) files  were downloaded in xml. The BeautifulSoup library was used to extract the abstract, title, and PMID. All files were stored in jsonl format."),
+        P(B("Filters Applied: "), "Multiple filters are used here after manually verifying output of all the filters as suggested by peS2o dataset."),
+        Ul(
+            Li("Minimum Word Count Filter: PMC documents with less than 100 words (not inclusive) are discarded; PMA documents less than 20 words are discarded", style = "margin-bottom: -3px"),
+            Li("Language Filter: English only", style = "margin-bottom: -3px"),
+            Li("Frequency Filter: The most frequent word in the paper consists of alpha characters only, and it appears in less than 7.5% of the document. Words are obtained by splitting the text on whitespace. This filter was not used for PMA", style = "margin-bottom: -3px"),
+            Li("Unigram Log Probability Threshold: -20", style = "margin-bottom: -3px"),
         ),
         table_div_med,
         Details(
@@ -677,19 +667,19 @@ filtering_process = Div(
         H3("Phil Papers"),
         P("Papers from the PhilPapers database, a comprehensive index and bibliography of philosophy research maintained by the Center for Digital Philosophy at the University of Western Ontario."),
         P(B("Download and Extraction: "), "Original PDF files download from", A("https://philarchive.org/oai.pl", href="https://philarchive.org/oai.pl"), ". All available PDF's were downloaded. Each PDF was converted to text using java", D_code("-jar ../philpapers_resources/src/pdfbox-app-2.0.21.jar ExtractText {f0} {FOUT.name}", language="java"), ". After converting to text formatting, a language was detected and added using the langdetect (citation needed) library."),
-        H4("Filtering"), 
+        P(B("Filters Applied: ")),
         Ul(
-            Li(P(B("Hyphenation Removal:"), D_code("end-of", language="python"), " becomes ", D_code("end of", language="python")), style = "margin-bottom: 2px"),
-            Li(P(B("Newline Filtering:"), D_code("This is/na sentence.", language="python"), " becomes ", D_code("This is a sentence.", language="python")), style = "margin-bottom: 2px"),
-            Li(P(B("Header/Footer Filtering:"), D_code("(c) 2023 Company Name.", language="python"), " is removed ",), style = "margin-bottom: 2px"),
-            Li(P(B("Double Whitespace Filtering:"), D_code("This  is  a  test.", language="python"), " becomes ", D_code("This is a test.", language="python")), style = "margin-bottom: 2px"),
-            Li(P(B("Mean Line Length Check: "), "removes paragraphs with an average line length of < 2.0"), style = "margin-bottom: 2px"),
-            Li(P(B("CID Percentage Filter: "), "removes LaTex heavy paragraphs that contain over 10% “CID” font artifacts."), style = "margin-bottom: 2px"),
-            Li(P(B("Letterness Filter: "), "discards paragraphs with a low proportion of letters"), style = "margin-bottom: 2px"),
-            Li(P(B("Removing Leading/Trailing Numbers: "), "removes numbers at the start or end of paragraphs. ", D_code("1 This is a sentence.", language="python"), " becomes ", D_code("This is a sentence.", language="python")), style = "margin-bottom: 2px"),
-            Li(P(B("Fixing Unicode Issues: "), "fixes Unicode issues."), style = "margin-bottom: 2px"),
-            Li(P(B("Combining Diacritics Correction: "), D_code("a'", language="python"), " becomes ", D_code("å", language="python")), style = "margin-bottom: 2px"),
-            Li(P(B("Unigram Log Probability: "), "the document must have higher than -20 average unigram log probability."), style = "margin-bottom: 2px"),
+            Li(P(B("Hyphenation Removal:"), D_code("end-of", language="python"), " becomes ", D_code("end of", language="python")), style = "margin-bottom: -3px"),
+            Li(P(B("Newline Filtering:"), D_code("This is/na sentence.", language="python"), " becomes ", D_code("This is a sentence.", language="python")), style = "margin-bottom: -3px"),
+            Li(P(B("Header/Footer Filtering:"), D_code("(c) 2023 Company Name.", language="python"), " is removed ",), style = "margin-bottom: -3px"),
+            Li(P(B("Double Whitespace Filtering:"), D_code("This  is  a  test.", language="python"), " becomes ", D_code("This is a test.", language="python")), style = "margin-bottom: -3px"),
+            Li(P(B("Mean Line Length Check: "), "removes paragraphs with an average line length of < 2.0"), style = "margin-bottom: -3px"),
+            Li(P(B("CID Percentage Filter: "), "removes LaTex heavy paragraphs that contain over 10% “CID” font artifacts."), style = "margin-bottom: -3px"),
+            Li(P(B("Letterness Filter: "), "discards paragraphs with a low proportion of letters"), style = "margin-bottom: -3px"),
+            Li(P(B("Removing Leading/Trailing Numbers: "), "removes numbers at the start or end of paragraphs. ", D_code("1 This is a sentence.", language="python"), " becomes ", D_code("This is a sentence.", language="python")), style = "margin-bottom: -3px"),
+            Li(P(B("Fixing Unicode Issues: "), "fixes Unicode issues."), style = "margin-bottom: -3px"),
+            Li(P(B("Combining Diacritics Correction: "), D_code("a'", language="python"), " becomes ", D_code("å", language="python")), style = "margin-bottom: -3px"),
+            Li(P(B("Unigram Log Probability: "), "the document must have higher than -20 average unigram log probability."), style = "margin-bottom: -3px"),
         ),
         table_div_phil,
         Details(
@@ -712,8 +702,7 @@ filtering_process = Div(
         H3("Europarl"),
         P("A collection of multilingual parallel corpora of parliamentary debates from the European Parliament. This is a high-quality legacy dataset earlier used for translation tasks."),
         P(B("Download and Extraction: "), "Original dataset was downloaded from", A("http://www.statmt.org/europarl/v7/europarl.tgz", href="http://www.statmt.org/europarl/v7/europarl.tgz"),". The files were converted to jsonl lines for filtering."),
-        H4("Filtering"),
-        P("EuroParl was initially filtered during the download process. Documents with fewer than 200 characters were removed. The documents also contained HTML tags which were removed."),
+        P(B("Filters Applied: ") ,"EuroParl was initially filtered during the download process. Documents with fewer than 200 characters were removed. The documents also contained HTML tags which were removed."),
         D_code("""
         Raw single line in data: <P> Hi I am speaker
         After tag removal: P Hi I am speaker
@@ -724,27 +713,20 @@ filtering_process = Div(
         D_code("""
         def process_tag(original_tag):
             tag = original_tag.strip(">").strip("<")
-        
             # Skip empty tags
             if not tag:
                 return None
-        
             tagname = tag.split()[0]
-        
             # Skip paragraph, break, and chapter tags
             if tagname in ["P", "BRK", "CHAPTER", "/P"]:
                 return None
-        
             # For speaker tags, return the name
             if tagname == "SPEAKER":
                 soup = bs4.BeautifulSoup(original_tag, "html.parser")
                 name = soup.speaker["name"]
                 return name
-        
             # Raise a error here if there is a tag we don't know
                 raise ValueError(f"Unknown tag {tag}")
-        
-        
         """, style="block", language = "python"),
         table_div_up,
         Details(
@@ -768,11 +750,11 @@ filtering_process = Div(
         P("High-quality dialog-based dataset where user comments on the links as the head post aggregated by Y Combinator."),
         P(B("Download and Extraction: "), "The dataset was downloaded from the HackerNews repo here:", A("https://hacker-news.firebaseio.com/v0/item/", href="https://hacker-news.firebaseio.com/v0/item/"), ". The dataset was parsed using the Story ID. In this dataset each post is a story, and each reply is considered subsequent story. Story IDs were considered between ID 1 to 37500000.  The URL for all Story IDs was pinged. If that ID returned an error, the ID was removed. Each request was given a 2 second wait to account for network time."),
         P("The HackerNews dataset contains a vast amount of stories and is known for lively discussions. Due to the number of replies a story may contain, only longest comment thread for each story was sampled past level 3. All stories included the title (1st level) and all direct replies (2nd level). Replies to the replies (3rd level) are only included for X STORIES."),
-        H4("Filtering"),
+        P(B("Filters Applied: ")),
         Ul(
-            Li("Language Filter: English", style = "margin-bottom: 2px"),
-            Li("Minimum Word Count Filter: 10", style = "margin-bottom: 2px"),
-            Li("Unigram Log Probability Threshold: -20", style = "margin-bottom: 2px"),
+            Li("Language Filter: English", style = "margin-bottom: -3px"),
+            Li("Minimum Word Count Filter: 10", style = "margin-bottom: -3px"),
+            Li("Unigram Log Probability Threshold: -20", style = "margin-bottom: -3px"),
         ),
         table_div_hn,
         ),
@@ -782,11 +764,11 @@ filtering_process = Div(
         H3("USPTO"),
         P("Patent documents from the United States Patent and Trademark Office."),
         P(B("Download and Extraction: "), "Data was downloaded and extracted using tags from", A("https://bulkdata.uspto.gov/data/patent/grant/redbook/fulltext/", href="https://bulkdata.uspto.gov/data/patent/grant/redbook/fulltext/"),". There were three different formats that needed three different functions to download and extract the data based on year:", I("Pre_2002"), ", ", I("2002_to_2004"), " and",  I("post_2004"),". We used the exact code used in The Pile (citation needed)."),
-        H4("Filtering"),
-        Ol(
-            Li("Language Filter: English", style = "margin-bottom: 2px"),
-            Li("Minimum Word Count Filter: 50", style = "margin-bottom: 2px"),
-            Li("Unigram Log Probability", style = "margin-bottom: 2px"),
+        P(B("Filters Applied: ")),
+        Ul(
+            Li("Language Filter: English", style = "margin-bottom: -3px"),
+            Li("Minimum Word Count Filter: 50", style = "margin-bottom: -3px"),
+            Li("Unigram Log Probability", style = "margin-bottom: -3px"),
         ),
         table_div_uspto,
         ),
@@ -795,8 +777,7 @@ filtering_process = Div(
         Div(
         H3("FreeLaw"),
         P("Legal documents and court cases from various jurisdictions provided by US-registered non-profit firm Free Law Project. We have included data from CourtListener which included millions of legal opinions from federal and state courts."),
-        H4("Download and Extraction"),
-        P("The dataset was downloaded from:", A("https://storage.courtlistener.com/bulk-data/", href="https://storage.courtlistener.com/bulk-data/"), ". There are 19 CSV files which contain overlapping content. CSV files can contain content in multiple columns requiring a holistic extraction approach. Text was extracted from the following using html2text function. The block below shows how each text type was extracted."), 
+        P(B("Download and Extraction"), "The dataset was downloaded from:", A("https://storage.courtlistener.com/bulk-data/", href="https://storage.courtlistener.com/bulk-data/"), ". There are 19 CSV files which contain overlapping content. CSV files can contain content in multiple columns requiring a holistic extraction approach. Text was extracted from the following using html2text function. The block below shows how each text type was extracted."), 
         D_code("""
         ("html", html2text),
         ("html_lawbox", html2text),
@@ -807,16 +788,13 @@ filtering_process = Div(
         plain_text
         """, language ="SQL"),
         P("All content was downloaded leading to high number of documents filtered during local deduplication. Following The Pile, priorty was given to plain_text first, followed by the columns in the table in reverse order."),
-        H4("Filtering"),
-        Ol(
-            Li("Language Filter: English", style = "margin-bottom: 2px"),
-            Li("Minimum Word Count Filter: 50", style = "margin-bottom: 2px"),
-            Li("Unigram Log Probability", style = "margin-bottom: 2px"),
-        ),
-        H4("Local Deduplication Process"),
-        Ol(
-            Li("Local dedup was done within freelaw itself which removed 90%+ duplicates"),
+        P(B("Filters Applied: ")),
+        Ul(
+            Li("Language Filter: English", style = "margin-bottom: -3px"),
+            Li("Minimum Word Count Filter: 50", style = "margin-bottom: -3px"),
+            Li("Unigram Log Probability", style = "margin-bottom: -3px"),
         ),
+        P("Note: Local deduplication within FreeLaw itself removed 90%+ of the dataset as duplicate."),
         table_div_freelaw,
         Details(
             Summary("FreeLaw Filtering Examples"),
@@ -850,9 +828,9 @@ filtering_process = Div(
   8. Comment1:
   9. Comment2:
         """),
-        H4("Filtering"),
-        Ol(
-            Li("Minimum Word Count Filter: 10", style = "margin-bottom: 2px"),
+        P(B("Filters Applied: ")),
+        Ul(
+            Li("Minimum Word Count Filter: 10", style = "margin-bottom: -3px"),
         ),
         table_div_se,
         Details(
@@ -888,11 +866,11 @@ filtering_process = Div(
         def clean(x):
             return '\n'.join('* ' + line[4:] if line.startswith('===') else line[8:] for line in x.split('\n'))
         """, block="block", language="python" ),
-        H4("Filtering"),
-        Ol(
-            Li("Language Filter: English", style = "margin-bottom: 2px"),
-            Li("Minimum Word Count Filter: 10", style = "margin-bottom: 2px"),
-            Li("Unigram Log Probability", style = "margin-bottom: 2px"),
+        P(B("Filters Applied: ")),
+        Ul(
+            Li("Language Filter: English", style = "margin-bottom: -3px"),
+            Li("Minimum Word Count Filter: 10", style = "margin-bottom: -3px"),
+            Li("Unigram Log Probability", style = "margin-bottom: -3px"),
         ),
         table_div_uirc,   
         ),
@@ -905,13 +883,9 @@ filtering_process = Div(
         D_code(""" 
         Question: TEXT
         Answer: TEXT""", block="block", language="python"),
-        H4("Filtering"),
-        Ol(
-            Li("No filtering was applied to DM Math"),
-        ),
-        H4("Local Deduplication Process"),
-        Ol(
-            Li("None"),
+        P(B("Filters Applied: ")),
+        Ul(
+            Li("No filtering was applied to DM Math", style = "margin-bottom: -3px"),
         ),
         table_div_dmm,
         Details(
@@ -933,9 +907,9 @@ filtering_process = Div(
         Div(
         H3("PG-19"),
         P("A collection of books from Project Gutenberg, a digital library of public domain works. This contains all the books that were published before 1919."),
-        P(B("Download and Extraction: "), "The dataset was downloaded directly from Huggingface:", A("https://huggingface.co/datasets/deepmind/pg19", href="https://huggingface.co/datasets/deepmind/pg19"), "."),
-        H4("Filtering"),
-        Ol(
+        P(B("Download and Extraction: "), "The dataset was downloaded directly from Huggingface: ", A("https://huggingface.co/datasets/deepmind/pg19", href="https://huggingface.co/datasets/deepmind/pg19"), "."),
+        P(B("Filters Applied:"))
+        Ul(
             Li("Language Filter: English", style = "margin-bottom: -3px"),
             Li("Minimum Word Count Filter: 20", style = "margin-bottom: -3px"),
             Li("Unigram Log Probability: ", "-20", style = "margin-bottom: -3px"),