The immediate results might present:<\/p>\n
Suppose the intent is not that the telescope is with my friend, so veto \u201ctelescope with my friend\u201d with a right-click.<\/p>\n You can also affirm that \u201cwith my friend\u201d is a complete prepositional phrase within the intent.<\/p>\n You can also affirm that \u201cwith the telescope\u201d is a complete prepositional phrase within the intent.<\/p>\n Suppose the intent is not about a park with a telescope, so veto that \u201cpark with a telescope\u201d with a right-click.<\/p>\n This leaves on a single derivation as shown below.<\/p>\n Now suppose that we were in the park but the man was not.\u00a0 If we veto \u201cman in the park\u201d we will have vetoed the last parse.\u00a0 (Try it, if you wish.\u00a0 Then press the reset button on the toolbar.)<\/p>\n Instead we can affirm that \u201cin the park\u201d is a complete prepositional phrase within the intent and reparse using the magnifying glass button on the toolbar (all of which is documented with tool-tips displayed upon hovering). and elsewhere at http:\/\/linguist.haleyai.com<\/a>.<\/p>\n In a default configuration, reparsing will ask if we want to use the parts of speech implied by our disambiguation thus far, such as follows:<\/p>\n The lack of a check on \u201csaw\u201d is beyond the intent of this article, but feel free to select it.\u00a0 Doing so has no practical effect in this case.<\/p>\n Next, because we have affirmed some parse tree structure, we can use that structure to constrain the parses we obtain on the next iteration.<\/p>\n This \u201cBracketing Dialog\u201d simply reflects the constraints on parse tree structure that we have specified.<\/p>\n Any fragment of text can be constrained to remain contiguous during parsing using this dialog.\u00a0 But be careful with it!\u00a0 You can easily choose constraints that are not grammatical, in which case you will find no results!<\/p>\n In most cases, this dialog is used to continue refining a parse or to start over, perhaps because of an error or uncertainty beforehand.<\/p>\n To use what you\u2019ve specified by affirming parse tree structure, press the OK button.\u00a0 To proceed to reparse without constraining parse structure toggle the selections and then press OK.\u00a0 The cancel button aborts the process of reparsing!<\/p>\n At this point, you will receive this dialog.<\/p>\n To get to our desired intent, veto \u201cman in the park\u201d to arrive at:<\/p>\n There is a little more grammatical disambiguation that can be done on the clauses tab.\u00a0 For example, whether \u201csaw\u201d is used in past or present tense!<\/p>\n Disambiguating Difficult Cases<\/p>\n The general guidelines for disambiguation have several aspects:<\/p>\n \u00b7 <\/span><\/span><\/span>reducing the degree of grammaticality will result in higher degrees of ambiguity (i.e., more time to parse and more resulting parses)<\/p>\n \u00b7 <\/span><\/span><\/span>the ambiguity of a sentences goes up significantly faster than its length<\/p>\n \u00b7 <\/span><\/span><\/span>certain words are highly ambiguous, and their ambiguities may multiply in the worst case<\/p>\n KnowBuddy™<\/span> and the Linguist™<\/span> addresses each of these so that correct parses can be obtained without extraordinary effort for longer sentences than most systems can handle.<\/p>\n The parsing tab of the settings dialog allows for setting the degree of grammaticality used in parsing.<\/p>\n \u00b7 <\/span><\/span><\/span>parse formal text results in the lowest number of parses<\/p>\n \u00b7 <\/span><\/span><\/span>parsing informal text introduces more ambiguity<\/p>\n \u00b7 <\/span><\/span><\/span>parsing fragmented text increases ambiguity significantly<\/p>\n \u00b7 <\/span><\/span><\/span>parsing robustly results in the highest degree of ambiguity<\/p>\n By ambiguity we are referring to the number of derivations (parses) that may be found for a piece of text.<\/p>\n Generally, parsing text formally should be used and then use parsing of fragments as necessary. <\/span>In some cases, however, robust parsing is necessary.<\/p>\n The parsing tab of the settings dialog may be used to adjust the amount of time parsing may take as well as the maximum number of parses that may be obtained.<\/p>\n Generally, these numbers should be in balance and reflect the degree of grammaticality with which parsing is being performed.<\/p>\n For example, to quickly work through relatively short, grammatical text from textbooks, parsing with formal grammaticality in less than 30 seconds with up to 100 derivations should be fine. <\/span>For less grammatical or longer sentences, 30 seconds for 2 to 4 hundred parses may be appropriate. <\/span>For longer sentences, a minute or more for perhaps 800 parses may seem necessary.<\/p>\n There are facilities in the Linguist that allow even loose grammaticality of long sentences to result in precise parsing, however.<\/p>\n It is not unusual to find no parses for a sentence.<\/p>\n In such cases, at least one of a few possibilities must be the case:<\/p>\n \u00b7 <\/span><\/span><\/span>the sentence is not grammatical, or some word or expression is used in a way that the parser does not handle<\/p>\n \u00b7 <\/span><\/span><\/span>there is too much ambiguity and parsing restrictions are set too low<\/p>\n o <\/span><\/span><\/span>e.g., the parser \u201ctimes out\u201d after finding a great many partial derivations without yet finding a full parse<\/p>\n o <\/span><\/span><\/span>e.g., the parse reports \u201cout of memory\u201d or \u201cedge limit exceeded\u201d before it finds any parses<\/p>\n \u00b7 <\/span><\/span><\/span>there are unknown words in the sentence and the parser setting to clarify unknown words has been turned off<\/p>\n o <\/span><\/span><\/span>note that applications may allow parsing to guess the part of speech of unknown words<\/p>\n o <\/span><\/span><\/span>also note that applications may provide part of speech information for known or unknown words<\/p>\n \u00b7 <\/span><\/span><\/span>there are some words in the sentence that have 2 or more parts of speech<\/p>\n The following sections address all the above except for unknown words, which are addressed elsewhere.<\/p>\n This section addresses 2 situations:<\/p>\n 1. <\/span><\/span><\/span>natural language processing is returning no parses due to exceeding one of the limits specified in the parsing tab of the settings dialog<\/p>\n 2. <\/span><\/span><\/span>natural language processing is returning no parses without exceeding any of those limits<\/p>\n In the 2nd<\/sup> case, there is either an unknown word or the sentence is not grammatical given the \u201cDegree of Grammaticality\u201d settings discussed above.<\/p>\n The 1st<\/sup> case above can arise for the combined reasons of too much ambiguity given the parser settings and a lack of grammaticality. <\/span><\/p>\n By addressing too much ambiguity as discussed here, the latter case, if applicable, becomes evident.<\/p>\n As discussed in the section \u201cDegree of Grammaticality\u201d above, ambiguity increases as the degree of grammaticality is relaxed from formal to informal or fragmented text.<\/p>\n Even parsing formal text, however, will not avoid the problem of obtaining no parses for some sentences. <\/span><\/p>\n In such cases, some combination of the following must be the case:<\/p>\n \u00b7 <\/span><\/span><\/span>the sentence is relatively long given the parser settings<\/p>\n \u00b7 <\/span><\/span><\/span>grammatical ambiguity is high given the parser settings<\/p>\n \u00b7 <\/span><\/span><\/span>lexical ambiguity is causing problems<\/p>\n Lexical ambiguity is easily handled as discussed below and elsewhere (e.g., with regard to parts of speech disambiguation).<\/p>\n For any degree of grammaticality, the ambiguity of a sentence increases more quickly than its length. <\/span>Thus, the first two bullets above are not independent but directly related.<\/p>\n For any given sentence, the number of possible parses increases faster than its length. <\/span>The reason for this is that the parser does not understand the text as person might.<\/p>\n Parsing considers plausible interpretations of words, phrases, and clauses and their composition into compound phrases and sentences. <\/span>This may result in combinatorial explosions.<\/p>\n Statistics help order parsing activity and resulting parses in a way that usually makes more sense to human beings who have common sense, but the parser itself knows nothing about meaning.<\/p>\n So, when the parser sees \u201cCAN\u201d, it must consider the possibility of that word being a noun for a metal container, a modal verb for ability or possibility, or an abbreviation for Canada.<\/p>\n \u00b7 <\/span><\/span><\/span>N possibilities for a single word in the sentence can, in the worst case, increase ambiguity by multiplication with N (although the impact is typically lower due to grammatical constraints).<\/p>\n \u00b7 <\/span><\/span><\/span>Note that applications can constrain the parts of speech considered in a variety of ways, including part of speech tagging and limiting the lexicon (e.g., not including \u201cCAN\u201d as an abbreviation).<\/p>\n Such ambiguity is \u201clexical\u201d in that it has to do with words not syntax or grammar. <\/span>The following are more specifically grammatical.<\/p>\n When the parser comes across a prepositional phrase, it may be plausible for that prepositional phrase to complement any phrase that came before it. <\/span><\/p>\n \u00b7 <\/span><\/span><\/span>\u201cI saw the man\u201d has some lexical ambiguity but no grammatical ambiguity. <\/span>(What kind of saw are you using?)<\/p>\n \u00b7 <\/span><\/span><\/span>\u201cI saw the man in the park\u201d has some grammatical ambiguity. (Where you in the park when you saw him or was the man in the park?)<\/p>\n \u00b7 <\/span><\/span><\/span>\u201cI saw the man in the park with the telescope\u201d has 4 times as much ambiguity. <\/span>(The parser does not know that you do not saw with a telescope!)<\/p>\n When the parser comes across disjunctions or conjunctions (which connect 2 or more phrases or clauses) there is frequent and possibly extreme ambiguity in what phrases or clauses it connects.<\/p>\n \u00b7 <\/span><\/span><\/span>\u201cWalk or chew gum\u201d is grammatically ambiguous. <\/span>(The parser does not know whether gum can be walked.)<\/p>\n \u00b7 <\/span><\/span><\/span>\u201cWalk or chew gum while listening\u201d is perhaps twice as ambiguous. <\/span>(Should you only listen while chewing or can you walk while listening?)<\/p>\n Another common but less pervasive source of ambiguity is compound words comprising more than 2 words.<\/p>\n \u00b7 <\/span><\/span><\/span>The parser does not know about carbon, steel, or bearings, so it considers \u201ccarbon steel bearings\u201d in the same way it would \u201csteel carbon bearings\u201d.<\/p>\n The foregoing are the primary sources of grammatical ambiguity. <\/span>Each of them is easily addressed using KnowBuddy, as described below.<\/p>\n Anyone can learn a lot about ambiguity using KnowBuddy!<\/p>\n 1. <\/span><\/span><\/span>Start KnowBuddy and log on, if necessary.[1]<\/span><\/span><\/span><\/span><\/a><\/p>\n 2. <\/span><\/span><\/span>Open the \u201cpublic\u201d knowledge-base.[2]<\/span><\/span><\/span><\/span><\/a><\/p>\n 3. <\/span><\/span><\/span>Double-click on a short sentence (up to 10 words long).<\/p>\n 4. <\/span><\/span><\/span>Hit the reset button on the toolbar of the analysis form that opens.[3]<\/span><\/span><\/span><\/span><\/a><\/p>\n Unless you have settings on the \u201canalysis\u201d tab of the Settings Dialog configured otherwise, you will be presented with disambiguation dialogs containing parse trees and parts of speech.<\/p>\n The following assumes you have seen and these dialogs, read about them in documentation provided elsewhere, and used them as encouraged above.<\/p>\n Other documentation addresses the various parts of speech and dialogs pertaining to parts of speech that are presented by KnowBuddy and the Linguist. <\/span>Here we focus on typically problematic sources of lexical ambiguity.<\/p>\n Assuming we are using the system with the following configuration (not all of which can be adjusted in the settings dialog):<\/p>\n \u00b7 <\/span><\/span><\/span>including Roman numerals (integers written using the letters I, V, X, L, C, D, M in either case)<\/p>\n \u00b7 <\/span><\/span><\/span>including uppercase letters as potentially proper (e.g., \u201cB\u201d may refer to some labeled item)<\/p>\n \u00b7 <\/span><\/span><\/span>supporting the parsing of dates (e.g., integers up to 31 may designate days of the month and integers up to 12 may designate months of the year)<\/p>\n \u00b7 <\/span><\/span><\/span>supported the parsing of times of day (e.g., integers up to 12 may designate hours of the day and integers up to 59 may designate minutes of an hour or seconds of a minute)<\/p>\n The following shows the result of parsing in a certain configuration where in under 10 seconds over 800 parses were obtained for the following sentence:<\/p>\n \u00b7 <\/span><\/span><\/span>I can mean one as in the roman numeral I or 1.<\/p>\n The problem in this trivial sentence is a combination of annoying lexical ambiguity and grammatical ambiguity.<\/p>\n The lexical ambiguities include:<\/p>\n \u00b7 <\/span><\/span><\/span>The 1st<\/sup> \u2018I\u2019 can be either a pronoun or the roman numeral for one.<\/p>\n \u00b7 <\/span><\/span><\/span>The \u2018one\u2019 can be either a pronoun, a modal auxiliary (correct), or a synonym of \u201cfire\u201d. <\/span>(Something can be a mean one!)<\/p>\n \u00b7 <\/span><\/span><\/span>The 2nd<\/sup> \u2018I\u2019 has the additional ambiguity that it could be a reference to section \u2018I\u2019 or the letter \u2018i\u2019 itself.[4]<\/span><\/span><\/span><\/span><\/a><\/p>\n \u00b7 <\/span><\/span><\/span>the \u20181\u2019 can be either the integer 1 or the day of the month numbered 1 or the hour of the day numbered 1.<\/p>\n Of course, we can apply statistical part of speech tagging to inform the parser that:<\/p>\n \u00b7 <\/span><\/span><\/span>The 1st<\/sup> \u2018I\u2019 is a pronoun.<\/p>\n \u00b7 <\/span><\/span><\/span>The \u2018one\u2019 is an integer or cardinality.<\/p>\n![\"\"](\"http:\/\/haleyai.com\/wordpress\/wp-content\/uploads\/image001.jpg\" "\\"image001\\"")
<\/a><\/p>\n<\/a><\/p>\n<\/a><\/p>\n<\/a><\/p>\n<\/a><\/p>\n<\/a><\/p>\n\n
<\/a><\/p>\n<\/a><\/p>\nDegree of Grammaticality<\/h2>\n
When there is too much Ambiguity<\/h2>\n
When there are no parses<\/h2>\n
Impact of Ambiguity<\/h2>\n
Sources of Ambiguity<\/h2>\n
Understanding Ambiguity<\/h2>\n
Ambiguity in Practice<\/h2>\n
Pathological Cases of Lexical Ambiguity<\/h2>\n
<\/span><\/p>\n