CN102200970A - Method and device for generating whole sentence in Chinese character input process - Google Patents

Method and device for generating whole sentence in Chinese character input process Download PDF

Info

Publication number
CN102200970A
CN102200970A CN2010101385319A CN201010138531A CN102200970A CN 102200970 A CN102200970 A CN 102200970A CN 2010101385319 A CN2010101385319 A CN 2010101385319A CN 201010138531 A CN201010138531 A CN 201010138531A CN 102200970 A CN102200970 A CN 102200970A
Authority
CN
China
Prior art keywords
name
dictionary
sentence
syllable
syllables
Prior art date
Application number
CN2010101385319A
Other languages
Chinese (zh)
Inventor
张会鹏
龚云华
Original Assignee
腾讯科技(深圳)有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 腾讯科技(深圳)有限公司 filed Critical 腾讯科技(深圳)有限公司
Priority to CN2010101385319A priority Critical patent/CN102200970A/en
Publication of CN102200970A publication Critical patent/CN102200970A/en

Links

Abstract

The invention discloses a method for generating a whole sentence in Chinese character input process. The method comprises the following steps of: dividing syllables of a pinyin string which is input by a user, and comparing each divided syllable with a preset family-name dictionary and a preset first-name dictionary; and when one or two continuous syllables in the pinyin string are matched and positioned in the family-name dictionary, and the one or two continuous syllables which are subsequently adjacent to the matched syllables are positioned in the first-name dictionary, providing personal names corresponding to a family-name syllable and a first-name syllable which are matched and using the personal names as candidate words for generation of the whole sentence, and displaying a generated result of the whole sentence by an interface. The invention also provides a device for generating the whole sentence in the Chinese character input process. By the method and the device, the probability of accurate prediction of the whole sentence when the pinyin string input by the user comprises personal-name pinyin is improved, so a better input experience is provided for the user.

Description

一种中文输入过程中的整句生成方法和装置 Sentence generating method and apparatus for Chinese character input process

技术领域 FIELD

[0001] 本发明涉及计算机的中文输入技术,尤其涉及一种中文输入过程中的整句生成方法和装置。 [0001] The present invention relates to computer input Chinese, particularly, to a method and apparatus for generating sentence Chinese input process.

背景技术 Background technique

[0002] 电脑使用者通常使用键盘中文输入软件将汉字输入到电脑,其中,键盘中文输入软件就是利用键盘,并根据一定的编码规则来输入汉字的一种方法。 [0002] Computer users usually use the keyboard input Chinese character input to the computer software, wherein the software is to use the Chinese input keyboard keypad, and a method for inputting Chinese characters according to certain encoding rules. 目前,汉字编码规则有数百种,在电脑上已经运行的有几十种,作为一种图形文字,汉字是由字的音、形、义来共同表达的,汉字输入的编码方法基本上都是采用将音、形、义与特定的键相联系,再根据不同汉字进行组合来完成汉字的输入的。 Currently, there are hundreds of Chinese character encoding rules, there are dozens on the computer has been running as a graphical text, kanji characters by sound, form and meaning to the expression of common, coding Chinese character input method basically the tone is the use, form and meaning associated with a particular key, and then combined in accordance with the different characters to complete the input of Chinese characters.

[0003] 目前主流的中文输入法都带有整句输入的功能,例如:用户需要输入“我们是未来的接班人”这句话时,在输入法软件中连续输入拼音串即可得到正确的整句结果,如下表所示: [0003] The current mainstream of Chinese input method sentence input function with, for example: Users need to enter the "We are the future successor" these words, in the input method Pinyin input software consecutive string to get the right sentence result, as shown in the following table:

[0004] [0004]

Figure CN102200970AD00041

[0005] 表1 [0005] TABLE 1

[0006] 但是,并不是输入的每一个拼音串都能得到用户想要的结果,例如:用户需要输入“这首诗是李白写的”这句话时,通过在输入法软件中连续输入拼音串会得到如下表所示的 [0006] However, each string is not a phonetic input of the user can get the results you want, for example: Users need to enter "This poem was written by Li Bai," these words, by continuous input Pinyin input method software string will be shown in the following table

结果: result:

[0007] [0007]

Figure CN102200970AD00042

[0008] 表2 [0008] TABLE 2

[0009] 出现上述问题的主要原因是:一个汉语拼音串通常都对应着多个汉字组词方案, 即汉字与拼音不是一一映射的,这导致了拼音输入法生成的结果带有不确定性。 Mainly [0009] The problems mentioned above are: Pinyin string is usually a plurality of Chinese characters corresponding to the program group of words, i.e., the phonetic characters is not one mapping, which results in the generation of the Pinyin input method with uncertain results . 为了实现整句输入,目前比较常用的整句生成算法是最大概率法,其基本原理是:在用户输入的一串拼音串中存在多种候选词的组合方式时,首先找出该拼音串中出现的所有候选词,然后在这些候选词的组合中找到一个概率最大的组合方式作为最后的整句生成结果。 In order to achieve sentence input, the current commonly used sentence generation algorithm is the maximum probability method, the basic principle is: the presence of various combinations of candidate words in a string of phonetic string input by the user, first find the Pinyin string All candidate word appears, and then find a probability largest combination in a combination of these candidate words as the last sentence to generate results.

[0010] 正如上述的最大概率法所描述,由于人名是不断变化的,不符合常规的组词规律, 而且人名也不是频繁输入的词语,这导致人名作为候选词出现的概率较低,从而整句生成的结果通常不易出现人名;而当用户输入的拼音串中确实需要出现人名时,生成的结果通常都不是用户所需,用户必须通过不断的翻页选词操作才能得到所需的整句结果,这显然影响了用户的输入体验。 [0010] As the above described maximum probability method, since the name is constantly changing, does not meet the rule of a conventional set of words, and word names are frequently not entered, which results in names as candidate word occurrence probability is low, so that the whole the results generated sentence is usually less prone names; and when the Pinyin string entered by the user does need names appear, the results generated are usually not required for the user, the user must obtain the required sentence through continuous operations to flip the choice of words the results, which obviously affects the user input experience. [0011] 由此可以看出,根据现有技术,目前的中文输入法将人名参与整句生成时的准确率不高,从而当用户输入的拼音串包含人名时,整句候选结果必然就无法满足用户所需,影响了用户的输入体验。 [0011] It can be seen, according to the prior art, the current input method of Chinese names participation accuracy when sentence generation is not high, so that when the user input comprises a Pinyin string names, sentence candidate can not be the inevitable result of meet user requirements, affecting the user input experience.

发明内容 SUMMARY

[0012] 有鉴于此,本发明的主要目的在于提供一种中文输入过程中的整句生成方法和装置,以解决现有技术将人名参与整句生成时的准确率不高的问题。 [0012] In view of this, the main object of the present invention to provide a method and apparatus for generating sentence Chinese input process of the prior art to solve the names participation accuracy when sentence generation is not high.

[0013] 为达到上述目的,本发明的技术方案是这样实现的: [0013] To achieve the above object, the technical solution of the present invention is implemented as follows:

[0014] 本发明提供了一种中文输入过程中的整句生成方法,该方法包括: [0014] The present invention provides a sentence generating method of Chinese character input process, the method comprising:

[0015] 对用户输入的拼音串进行音节划分,并将划分的各个音节与预设的姓氏字典和名字字典进行比较; [0015] The user input Pinyin syllable string is divided, and the divided syllables with a preset name and last name dictionary dictionary;

[0016] 当匹配到所述拼音串中存在一个或两个连续的音节位于所述姓氏字典中,且匹配音节后续紧邻的一个或两个连续的音节位于所述名字字典中时,将所匹配的姓氏音节和名字音节对应的人名作为候选词参与整句的生成,并将所述整句生成的结果通过界面显示。 [0016] When a match to the presence of one or Pinyin string of syllables in said two consecutive last name dictionary, and matching a syllable or syllables two successive located immediately subsequent to the name of the dictionary, the matched results name syllables and syllable surname name corresponding candidate word involved in generating a sentence, and the sentence generated by the display interface.

[0017] 所述姓氏字典和名字字典由服务器提供,客户端从服务器下载。 The [0017] last name dictionary and the dictionary name provided by the server, the client downloaded from the server.

[0018] 所述姓氏字典中包括收集的姓氏,以及每个姓氏的出现概率和对应的拼音;所述名字字典中包括收集的名字,以及每个名字的出现概率和对应的拼音。 [0018] Last name dictionary includes the collection of the last name, and the probability of occurrence of each family name and the corresponding pinyin; the name dictionary includes a collection of names, and the probability of occurrence and the corresponding pinyin each name.

[0019] 所述将所匹配的姓氏音节和名字音节对应的人名作为候选词参与整句的生成,具体为: [0019] The matching of the surnames of syllables and syllable corresponding names as name involved in generating candidate word sentence, specifically:

[0020] 根据所匹配的姓氏音节的出现概率和名字音节的出现概率,计算出对应人名作为候选词的综合出现概率; [0020] The probability of occurrence of the matched name and surname syllable syllable probability is calculated as a candidate word corresponding to the name of the integrated probability of occurrence;

[0021] 在生成整句时,按照预设条件选择综合出现概率较大的一个或多个人名作为候选词参与整句的生成。 [0021] When generating sentence, under a predetermined condition to select a larger overall probability of one or more people involved in generating name as a candidate word of the sentence.

[0022] 所述将划分的各个音节与预设的姓氏字典和名字字典进行比较,进一步包括: [0022] The division of the syllables with a preset name and last name dictionary dictionary, further comprising:

[0023] 将拼音串划分出的各个音节与所述姓氏字典进行比较,在匹配到所述拼音串中存在一个或两个连续的音节位于所述姓氏字典时,将所述匹配音节后续紧邻的一个音节与所述名字字典进行比较,判断所述紧邻的一个音节是否位于名字字典中,如果位于,则匹配成功;如果不位于,则将所述匹配音节后续紧邻的两个音节与所述名字字典进行比较,判断所述紧邻的两个音节是否位于名字字典中,如果位于,则匹配成功,否则匹配失败。 [0023] Pinyin syllables is divided string is compared with the last name dictionary, or the presence of a continuous two syllables in said last name dictionary matched to the Pinyin string, the syllable matched the immediately subsequent a syllable is compared with the name of the dictionary, a determination of the syllable is located immediately name dictionary, if located, the matching is successful; if not located, then the two syllables of the names match syllable immediately subsequent dictionary, and determines whether or not the two syllables is located next to the name of the dictionary, if located, the match is successful, otherwise the match fails.

[0024] 该方法进一步包括:在匹配失败时,选择所述音节对应的普通候选词参与整句的生成。 [0024] The method further comprising: when the matching fails, the selected candidate word syllable ordinary involved in generating the corresponding sentence of.

[0025] 本发明还提供了一种中文输入过程中的整句生成装置,该装置包括: [0025] The present invention further provides an apparatus for generating a sentence in the Chinese character input process, the apparatus comprising:

[0026] 音节划分比较模块,用于对用户输入的拼音串进行音节划分,并将划分的各个音节与预设的姓氏字典和名字字典进行比较; [0026] syllabication comparison module for user input Pinyin string is divided into syllables, syllables and divides each compared with a preset first name and last name dictionary dictionary;

[0027] 整句生成模块,用于根据所述音节划分比较模块的比较结果,在匹配到所述拼音串中存在一个或两个连续的音节位于所述姓氏字典中,且匹配音节后续紧邻的一个或两个连续的音节位于所述名字字典中时,将所匹配的姓氏音节和名字音节对应的人名作为候选词参与整句的生成; [0027] sentence generation module for dividing the comparison result of the comparison module in accordance with the syllable, or the presence of a continuous two syllables in the Pinyin string to match in the dictionary is in the last name, and the immediately subsequent match syllables when one or two successive syllables in said name dictionary, the matched name and surname syllables corresponding syllable as the name of the candidate words involved in generating sentence;

[0028] 显示模块,用于将所述整句生成的结果通过界面显示。 [0028] The display module, the sentence for the result generated by the interface display. [0029] 所述姓氏字典和名字字典由服务器提供,整句生成装置所在的客户端从服务器下载。 [0029] The last name dictionary and the dictionary name provided by the server, where the sentence generation means from the server to download client.

[0030] 所述姓氏字典中包括收集的姓氏,以及每个姓氏的出现概率和对应的拼音;所述名字字典中包括收集的名字,以及每个名字的出现概率和对应的拼音。 [0030] Last name dictionary includes the collection of the last name, and the probability of occurrence of each family name and the corresponding pinyin; the name dictionary includes a collection of names, and the probability of occurrence and the corresponding pinyin each name.

[0031] 所述整句生成模块进一步包括: [0031] The sentence generation module further comprises:

[0032] 概率计算子模块,用于根据所匹配的姓氏音节的出现概率和名字音节的出现概率,计算出对应人名作为候选词的综合出现概率; [0032] probability calculating sub-module, according to the occurrence probability match the name and surname syllable syllable probability is calculated as a candidate word corresponding to the name of the integrated probability;

[0033] 整句生成子模块,用于在生成整句时,按照预设条件选择综合出现概率较大的一个或多个人名作为候选词参与整句的生成。 [0033] sentence generation sub-module for generating sentence, according to a preset condition selection overall probability of occurrence of one or more large personal name as a candidate involved in generating word of the sentence.

[0034] 所述音节划分比较模块进一步用于,将拼音串划分出的各个音节与所述姓氏字典进行比较,在匹配到所述拼音串中存在一个或两个连续的音节位于所述姓氏字典时,将所述匹配音节后续紧邻的一个音节与所述名字字典进行比较,判断所述紧邻的一个音节是否位于名字字典中,如果位于,则匹配成功;如果不位于,则将所述匹配音节后续紧邻的两个音节与所述名字字典进行比较,判断所述紧邻的两个音节是否位于名字字典中,如果位于, 则匹配成功,否则匹配失败。 [0034] The comparison module is further used for syllable division, the divided Pinyin string of syllables is compared with the last name dictionary, or the presence of a continuous two syllables in said dictionary matching surnames to the Pinyin string when the name matches a syllable dictionary and the immediately subsequent syllable, and determines whether the syllable immediately positioned name dictionary, if located, the matching is successful; if not located, then the match syllable follow immediately with two syllables of the name dictionary, and determines whether or not the two syllables is located next to the name of the dictionary, if located, the match is successful, otherwise the match fails.

[0035] 所述整句生成模块进一步用于,在匹配失败时,选择所述音节对应的普通候选词参与整句的生成。 [0035] The sentence generation module is further configured, when the match fails, the selected candidate word syllable ordinary involved in generating the corresponding sentence of.

[0036] 本发明所提供的一种中文输入过程中的整句生成方法和装置,对用户输入的拼音串进行音节划分,并将划分的各个音节与预设的姓氏字典和名字字典进行比较;当匹配到拼音串中存在一个或两个连续的音节位于姓氏字典中,且匹配音节后续紧邻的一个或两个连续的音节位于名字字典中时,将所匹配的姓氏音节和名字音节对应的人名作为候选词参与整句的生成,并将整句生成的结果通过界面显示。 [0036] A Chinese character input process of the present invention provides a sentence generating method and the apparatus, user input Pinyin syllable string is divided, and the divided syllables with a preset name and last name dictionary dictionary; when matched to a Pinyin string exists in two successive syllable or last name dictionary is located, and one or two consecutive matches on syllable is located at the immediately subsequent name dictionary, the matched name and surname syllables corresponding syllable name involved in generating candidate words as a result of the sentence, and sentence display generated by the interface. 通过本发明,提高了当用户输入拼音串中包含人名拼音时的整句预测正确的概率,从而给用户以更好的输入体验。 By the present invention improves the probability of correct history when the user inputs the Pinyin string names comprising the sentence prediction, and thus a better input to the user experience.

附图说明 BRIEF DESCRIPTION

[0037] 图1为本发明一种中文输入过程中的整句生成方法的流程图一; A flowchart of a method of generating a sentence [0037] Figure 1 is a disclosure of a Chinese character input process;

[0038] 图2为本发明一种中文输入过程中的整句生成方法的流程图二; [0038] FIG 2 is a flowchart of a method of generating two Chinese sentence input in the process of the invention;

[0039] 图3为本发明实施例中拼音串比较匹配的示意图; [0039] FIG. 3 is a schematic embodiment Pinyin strings match comparison embodiment of the invention;

[0040] 图4为本发明一种中文输入过程中的整句生成装置的结构示意图。 [0040] FIG. 4 is a schematic structure of a Chinese sentence generating means in the input process of the present invention.

具体实施方式 Detailed ways

[0041] 下面结合附图和具体实施例对本发明的技术方案进一步详细阐述。 Drawings and specific embodiments set forth in detail further aspect of the present invention [0041] The following binding.

[0042] 本发明提供的一种中文输入过程中的整句生成方法,旨在当用户输入的拼音串中包含人名拼音时,增加该人名作为候选词参与整句生成,以提高整句预测正确的概率。 [0042] A Chinese character input process of the present invention provides a method of generating a sentence, when the intended user Pinyin string names included in the input alphabet, the person's name as a candidate word increase participation sentence generation sentence correctly predicted to increase The probability. 如图1所示,本发明的整句生成方法主要包括以下步骤: 1, the sentence generating method of the present invention includes the following steps:

[0043] 步骤101,对用户输入的拼音串进行音节划分,并将划分的各个音节与预设的姓氏字典和名字字典进行比较。 [0043] Step 101, the Pinyin syllable string entered by the user is divided, and the divided syllables with a preset dictionary surname and first name dictionary.

[0044] 对于大部分中文人名,一般由姓和名组成,其中常见的姓名通常包括以下形式: [0044] For most Chinese names, generally consists of first and last names, of which common names usually include the following forms:

[0045] 1、单姓+ —个字,如:项羽;[0046] 2、单姓+第一个字+第二个字,如:李世民; [0045] 1, single surname + - words, such as: Xiang; [0046] 2, a single-word surname + first + second word, such as: the emperor;

[0047] 3、复姓+ —个字,如:诸葛亮; [0047] 3, hyphenated + - words, such as: Liang;

[0048] 4、复姓+第一个字+第二个字,如:司马相如。 [0048] 4, a hyphenated word + + CAP second word, such as: Simaxiangru.

[0049] 为描述方便,下面描述中的姓氏特指人名中的姓,如诸葛亮中的诸葛、李白中的李;名字则指诸葛亮中的亮、李白中的白;人名则指完整的姓名,如司马相如。 [0049] For ease of description, the following description refers specifically surname name Names, as the Zhuge Liang, Bai LI; Liang name refers to the light, the white Bai; full name refers to the name, The Simaxiangru.

[0050] 本发明中根据收集的诸多姓名生成一个姓氏字典和一个名字字典,姓氏字典中除了保存姓氏外,还保存包括每个姓氏对应的一个概率值和对应的拼音,该概率值用于表征姓氏的出现频率,也称出现概率;名字字典中除了保存名字外,还保存包括每个名字对应的一个概率值和对应的拼音,该概率值用于表征名字的出现频率,也称出现概率。 [0050] The present invention generates a name and a last name dictionary dictionary according to collect many first and last name dictionary stored in addition to the last name, but also save and comprising a corresponding probability value corresponding to the last name of each pinyin, the probability value for characterizing frequency of occurrence of the surname, also called probability; name dictionary stored in addition to the name, but also includes saved and a probability value corresponding to each name corresponding pinyin, which probability value is used to characterize the frequency of occurrence of the name, also called probability. 其中,姓氏和名字的收集可以采取多种方式,例如:将所有百家姓中的姓氏全部加入姓氏字典中,并将人口普查中统计的每个姓氏的出现频率作为对应各个姓氏的出现概率;或者,只将人口普查中统计的出现频率较高的若干个姓氏加入姓氏字典中,并将这些姓氏的出现频率作为对应各个姓氏的出现概率;同样的,也可以将人口普查中统计的出现频率较高的若干个名字加入名字字典中,并将这些名字的出现频率作为对应各个名字的出现概率。 Among them, first and last name of the collection can take a variety of ways, for example: All Family Names in the last name added to all surnames dictionary, the frequency of occurrence and census statistics for each surname surname as the probability of occurrence of each of correspondence; Alternatively, only the statistics appear census plurality of higher frequency dictionary surnames surnames added, and as the occurrence probability corresponding to the frequency of occurrence of each surname of these names; the same, may be the frequency of occurrence statistics census a higher number of names added to the name of the dictionary, and as the probability of occurrence of each name corresponds to the frequency of occurrence of these names. 本发明中也可以通过语料训练的方式来收集常用的姓氏和名字,并获取各自对应的出现概率。 The present invention may also be used to collect the first and last name by way of training corpus, and obtain respectively corresponding probability of occurrence. 本发明并不对姓氏和名字的收集方式进行限定。 The present invention is not collected embodiment first and last name is defined.

[0051] 这些最初的姓氏字典和名字字典可以由服务器提供,客户端从服务器上下载。 [0051] These initial last name and first name dictionary dictionary can be provided by the server, the client downloaded from the server. 当然,该姓氏字典和名字字典也支持后续更新,服务器上的姓氏字典和名字字典可以更新后重新发布,客户端下载更新;客户端上的姓氏字典和名字字典也可以支持主动更新,即用户可以对客户端上的姓氏字典和名字字典进行主动添加项、删除项等操作,其操作的结果还可以上报给服务器,由服务器同步更新。 Of course, the surname dictionary and name dictionary also supports subsequent updates, last name dictionary and the name of the dictionary on the server after can be updated re-release, client download updates; surname dictionary and names dictionary on the client also supports automatic update that the user can dictionary of surnames and names dictionary on the client actively add items, remove items and other operations, the results of its operations can also be reported to the server, synchronize updates from the server.

[0052] 在用户执行中文输入的过程中,首先对用户输入的拼音串进行音节划分,再将划分的各个音节与预设的姓氏字典和名字字典进行比较,通过比较判断用户输入的拼音串中是否包含人名拼音。 [0052] In the Chinese input process performed by the user, the first user input string of pinyin syllables is divided, then the divided syllables with a preset dictionary surname and first name dictionary, is determined by comparing the Pinyin string input by the user It contains names Pinyin. 具体的:将拼音串划分出的各个音节与姓氏字典进行比较,在匹配到拼音串中存在一个或两个连续的音节位于姓氏字典时,将匹配音节后续紧邻的一个音节与名字字典进行比较,判断紧邻的一个音节是否位于名字字典中,如果位于,则匹配成功;如果不位于,则将匹配音节后续紧邻的两个音节与名字字典进行比较,判断紧邻的两个音节是否位于名字字典中,如果位于,则匹配成功,否则匹配失败。 Specifically: Pinyin syllables with the surname string dictionary partitioned comparing, there is a continuous two syllables or last name dictionary is located, the name matches a syllable dictionary syllable immediately subsequent matching to compare the Pinyin string, judgment of a syllable is located next to the name of the dictionary, if located, the match is successful; if not located, the names of two syllables and syllables dictionary will immediately follow-up match, and determines whether or not located in close proximity to the two-syllable name dictionary, If located, the match is successful, otherwise the match fails.

[0053] 步骤102,当匹配到拼音串中存在一个或两个连续的音节位于姓氏字典中,且匹配音节后续紧邻的一个或两个连续的音节位于名字字典中时,将所匹配的姓氏音节和名字音节对应的人名作为候选词参与整句的生成,并将整句生成的结果通过界面显示。 When [0053] Step 102, the Pinyin string matching when there is one or two successive syllables located surname dictionary, and match one or two consecutive follow immediately positioned on syllable dictionary name, the surname syllable matched results syllables and corresponding names as name involved in generating candidate word sentence, and sentence display generated by the interface.

[0054] 本发明中可以根据所匹配的姓氏音节的出现概率和名字音节的出现概率,计算出对应人名作为候选词的综合出现概率,例如:通过姓氏音节的出现概率与名字音节的出现概率加权运算得到对应人名作为候选词的综合出现概率;然后在生成整句时,按照预设条件选择综合出现概率较大的一个或多个人名作为候选词参与整句的生成。 [0054] The present invention may probability of occurrence of the matched last name syllable probability and the name of syllables, calculate the corresponding name as the candidate word probability synthesis occurs, for example: by weighting the probability of occurrence probabilities and the name syllable surname syllable calculating a name to give the corresponding candidate word occurrence probability synthesis; then in generating the sentence, according to a preset condition selection probability of a larger integrated one or more people involved in generating candidate words as the name of the sentence. 举例如下:姓氏音节Al的出现概率为0. 05,名字音节Bl的出现概率为0. 03,那么加权运算得到对应人名(即人名A1+B1)作为候选词的综合出现概率为0.05 α+0.03 β,其中α为姓氏音节的加权值,β为名字音节的加权值,该加权值可以根据实际需要进行选取,α和β的取值可以相等;如果用户设定的预设条件为:允许选择名字作为相应音节对应的候选词所生成的整句不超过两个时,那么选择综合出现概率较大的两个人名作为候选词参与整句的生成。 For example as follows: Last Name syllable occurrence probability of Al is 0.05, the probability of occurrence of the name syllable Bl is 0.03, then the weighting operation to obtain the corresponding names (i.e. names A1 + B1) as a candidate word occurrence probability is integrated 0.05 α + 0.03 beta], where [alpha] is a weighting value surname syllables, syllables beta] is the weighted value of the name, the weight value can be selected according to actual needs, and [alpha] may be equal to the value of beta]; if the preset condition set by the user to: allow selection name as a candidate word corresponding syllable corresponding generated when the sentence does not exceed two, then the probability of occurrence of a large comprehensive selection of two names generated word sentence as a candidate to participate.

[0055] 此外在匹配失败时,需要选择各音节对应的普通候选词参与整句的生成。 [0055] Further when the matching fails, it is necessary to select a candidate word syllables ordinary involved in generating the corresponding sentence of. 也就是说,在对姓氏字典或名字字典的匹配失败时,从普通语料库中选择该音节对应的普通候选词参与整句的生成。 In other words, when matching names or surnames dictionary dictionary fails, select Normal candidate word corresponding to the syllable involved in generating the sentence from the general corpus.

[0056] 基于图1所示的整句生成方法,图2提供了一种文输入过程中的整句生成方法的具体实现流程,主要包括以下步骤: [0056] The method of generating a sentence based on FIG. 1, FIG. 2 provides a specific implementation of a method for generating a flow of sentence text input process, including the steps of:

[0057] 步骤201,在用户输入拼音串时,客户端进行音节的划分。 [0057] Step 201, when the user inputs the Pinyin string, the client into syllables.

[0058] 步骤202,判断音节划分后的拼音串中是否存在一个音节位于姓氏字典中的情况, 如果存在,执行步骤203 ;否则,执行步骤207。 [0058] Step 202, it is determined whether there is located a syllable dictionary surname Pinyin string after syllable division, if present, step 203; otherwise, step 207 is executed.

[0059] 将划分后的各个音节与预设的姓氏字典进行比较,即将划分后的各个音节与姓氏字典中的拼音进行一一匹配,以判断拼音串中是否有音节位于姓氏字典中。 [0059] After the respective syllable division for comparison with a preset dictionary surname, each syllable dictionary surname after coming one by one division phonetic matching, to determine whether there Pinyin syllables located surname string dictionary. 这里所说的匹配是指完全匹配,即只有在划分后的某个音节与姓氏字典中的某个拼音完全匹配时,才判断该音节位于姓氏字典中。 Here's match is an exact match, that is, only a phonetic dictionary in one syllable with the surname of the divided exactly match only when the judge syllable located surname dictionary. 该步骤只判断单姓的情况,先不考虑复姓的情况。 Analyzing the case of a single step only the name, is not considered the case of hyphenated. 步骤202的操作实际上就是为了判断用户输入的拼音串中是否包括单姓的姓氏拼音。 Procedure 202 is to actually determined phonetic alphabet input by the user whether the last name surname comprises a single string.

[0060] 步骤203,判断步骤202中匹配音节后续紧邻的一个音节是否位于名字字典中,如果位于,执行步骤208 ;否则,执行步骤204。 [0060] Step 203, it is determined in step 202 subsequent match syllable a syllable is located immediately name dictionary, if located, step 208 is executed; otherwise, step 204 is performed.

[0061] 将匹配音节后续紧邻的一个音节与名字字典进行比较,即将该紧邻的一个音节与名字字典中的拼音进行一一匹配,以判断名字字典中是否存在该紧邻的一个音节。 [0061] will match a syllable with the syllable name dictionary to compare the immediate follow-up, about to close one of the matches one by one syllable name phonetic dictionary to determine if there is a syllable of the name immediately dictionary. 这里所说的匹配也是指完全匹配,即只有在划分后的某个音节与名字字典中的某个拼音完全匹配时,才判断该音节位于名字字典中。 The match is also mentioned here refers to an exact match, that is, only when a Pinyin syllable in the name of a dictionary of the divided exactly match before the judge syllable name is located in the dictionary. 步骤203的操作实际上就是为了判断步骤202中单姓的姓氏拼音后紧邻的一个音节是否可以作为姓名中的字。 Procedure 203 is actually a family names to decision step 202 after the single name immediately whether a syllable as the name of the word.

[0062] 步骤204,判断步骤202中匹配音节后续紧邻的两个音节是否位于名字字典中,如果位于,执行步骤208 ;否则,执行步骤205。 [0062] Step 204, it is determined in step 202 subsequent match syllable two syllables is located immediately name dictionary, if located, step 208 is executed; otherwise, step 205 is executed.

[0063] 将匹配音节后续紧邻的两个音节与名字字典进行比较,即将该紧邻的两个音节与名字字典中的拼音进行一一匹配,以判断名字字典中是否存在该紧邻的两个音节。 [0063] will match the names of two syllables and syllables dictionary close follow-up were compared with two syllables of the name of the dictionary is about to close in spelling to match one by one, to determine whether the presence of two syllables of the name next to the dictionary. 这里所说的匹配也是指完全匹配,即只有在紧邻的两个音节与名字字典中的某个拼音完全匹配时,才判断该音节位于名字字典中。 The match is also mentioned here refers to an exact match, ie only two syllables with an alphabet in the dictionary next to the names of the exact match, it determines whether the syllable name is located in the dictionary. 步骤204的操作实际上就是为了判断步骤202中单姓的姓氏拼音后紧邻的两个音节是否可以作为姓名中的字。 Procedure 204 is actually a family names in order to determine in step 202 immediately after a single name if the two syllables of the word as the name.

[0064] 步骤205,判断是否检查了复姓,如果是,执行步骤207 ;否则,执行步骤206。 [0064] Step 205 determines whether the check hyphenated, if yes, perform step 207; otherwise, step 206 is performed.

[0065] 如果判断已经检查过了复姓,那么已经可以确定用户输入的拼音串中不包含人名拼音,可以执行步骤207 ;如果还没有检查过复姓,那么需要转到步骤206执行复姓的检查操作。 [0065] If a determination has been checked hyphenated, it can be readily determined phonetic string input by the user does not contain spelling names, may perform step 207; if it is not checked hyphenated, then goes to step 206 requires a check operation performed hyphenated.

[0066] 步骤206,判断拼音串中是否存在两个连续的音节位于姓氏字典中,如果是,执行步骤203 ;否则,执行步骤207。 [0066] Step 206, it is determined whether there is a Pinyin syllable string is located in two consecutive last name dictionary, if yes, perform step 203; otherwise, step 207 is executed.

[0067] 将划分后的每两个连续音节与预设的姓氏字典进行比较,即将划分后的每两个连续音节与姓氏字典中的复姓拼音进行一一匹配。 [0067] each syllable after two consecutive division with a preset dictionary of surnames, hyphenated spelling each of two consecutive syllables with the surname dictionary division after coming in one by one match.

[0068] 步骤202的操作是为了判断用户输入的拼音串中是否包括单姓的姓氏拼音,而步骤206的操作则是为了判断用户输入的拼音串中是否包括复姓的姓氏拼音。 Operation [0068] Step 202 is to determine whether the spelling of the name of family names includes a single string input by the user, and the operation of step 206 is to determine the Pinyin string input by the user is included in the family names hyphenated.

[0069] 步骤207,生成步骤202中的匹配音节所对应的普通候选词,执行完后转到步骤209。 [0069] Step 207, the generated match syllables in step 202 candidate word corresponding to the normal, proceeds to step 209 after the implementation.

[0070] 步骤208,根据所匹配的姓氏音节和名字音节生成对应的人名候选词,并根据所匹配的姓氏音节的出现概率和名字音节的出现概率,计算出对应人名作为候选词的综合出现概率。 [0070] Step 208 generates a name candidate words corresponding to in accordance with the matched surname syllables and names syllable, and in accordance with the probability occurrence of the matched last name syllable probability and the name of syllables, calculate the corresponding name as the candidate word probability synthesis occurs .

[0071] 步骤209,继续根据拼音串的音节划分结果生成其他的候选词。 [0071] Step 209, to continue to generate additional candidate words according to syllable division result Pinyin string.

[0072] 步骤210,按照预设条件选择综合出现概率较大的一个或多个人名作为候选词参与整句的生成,并将整句生成的结果通过界面显示。 [0072] Step 210, according to a preset condition selection probability of occurrence of a large integrated one or more people involved in generating candidate words as the name of the sentence and the sentence generated by the interface display results.

[0073] 例如,当用户设定的预设条件为:允许选择名字作为相应音节对应的候选词所生成的整句不超过两个时,选择综合出现概率较大的两个人名作为候选词参与整句的生成。 [0073] For example, when a preset condition set by the user to: allow selection of a sentence Name syllable corresponding to the respective candidate word generated by not more than two, the larger the probability of two names selected as the candidate words appear integrated participation generation of sentence.

[0074] 在以下的实施例中,当用户需要输入“这首诗是李白写的”这句话时,在输入法软件中连续输入拼音串:zhe' shou,sh,shi,li,bai,xie' de,那么根据前述图2所示的整句生成方法,其拼音串比较匹配的示意图如图3所示,通过这些操作即可得到如下表所示的显示结果, [0074] In the following embodiment, when a user needs to input "is a poem written by Bai" this sentence, the continuous input Pinyin string input in the software: zhe 'shou, sh, shi, li, bai, xie 'de, then the sentence in accordance with the generation method shown in FIG. 2, which match the Pinyin string comparison diagram shown in Figure 3, these operations can be obtained by the following table shows the result,

[0075] [0075]

Figure CN102200970AD00091
Figure CN102200970AD00092

[0076]表 3 [0076] TABLE 3

[0077] 通过这样的操作,当用户输入的拼音串中确实需要出现人名时,生成的整句结果就能够出现用户所需的人名。 [0077] By so doing, when the string appears phonetic name does require a user input, generating a result of the sentence can be required by the user name appears.

[0078] 对应上述中文输入过程中的整句生成方法,本发明还提供了一种中文输入过程中的整句生成装置,如图4所示,该装置主要包括:音节划分比较模块10、整句生成模块20和显示模块30。 [0078] corresponding to the above-described method of generating Chinese sentence in the input process, the present invention also provides an apparatus for generating a sentence in the Chinese character input process shown in Figure 4, the apparatus including: dividing the syllable comparison module 10, the whole sentence generation module 20 and the display module 30. 其中,音节划分比较模块10,用于对用户输入的拼音串进行音节划分,并将划分的各个音节与预设的姓氏字典和名字字典进行比较。 Wherein syllabication comparison module 10, a user input for Pinyin string is divided into syllables, syllables and divides each compared with a preset first name and last name dictionary dictionary. 姓氏字典和名字字典由服务器提供,整句生成装置所在的客户端从服务器下载。 Last name and first name dictionary dictionary provided by the server, the client device generates a sentence where downloaded from the server. 整句生成模块20,用于根据音节划分比较模块的比较结果,在匹配到拼音串中存在一个或两个连续的音节位于姓氏字典中,且匹配音节后续紧邻的一个或两个连续的音节位于名字字典中时,将所匹配的姓氏音节和名字音节对应的人名作为候选词参与整句的生成。 Sentence generation module 20, according to the comparison module syllabication comparison result, there is a matching string into phonetic syllables or two successive surname positioned dictionary, and match one or two successive located on syllable immediately subsequent when name dictionary, the matched name and surname syllables corresponding syllable as the name of the candidate words involved in generating sentence. 显示模块30,用于将整句生成的结果通过界面显 The display module 30 for generating a sentence by interfacial significant results

7J\ ο 7J \ ο

[0079] 整句生成模块20进一步包括:概率计算子模块21和整句生成子模块22。 [0079] sentence generation module 20 further comprising: a probability calculating module 21 and sub-sub-sentence generation module 22. 概率计算子模块21,用于根据所匹配的姓氏音节的出现概率和名字音节的出现概率,计算出对应人名作为候选词的综合出现概率。 Probability calculation sub-module 21, according to the probability of occurrence of the matched name and surname syllable syllable probability is calculated as a candidate word name corresponding to the probability of occurrence synthesis. 整句生成子模块22,用于在生成整句时,按照预设条件选择综合出现概率较大的一个或多个人名作为候选词参与整句的生成。 Sentence generating sub-module 22 for generating the sentence, under a predetermined condition to select a larger overall probability of one or more people involved in generating name as a candidate word of the sentence.

[0080] 此外,音节划分比较模块10进一步用于,将拼音串划分出的各个音节与姓氏字典进行比较,在匹配到拼音串中存在一个或两个连续的音节位于姓氏字典时,将匹配音节后续紧邻的一个音节与名字字典进行比较,判断紧邻的一个音节是否位于名字字典中,如果位于,则匹配成功;如果不位于,则将匹配音节后续紧邻的两个音节与名字字典进行比较, 判断紧邻的两个音节是否位于名字字典中,如果位于,则匹配成功,否则匹配失败。 [0080] Further, the comparison module 10 is further divided into syllables for the Pinyin syllables of the string divided comparing with the last name dictionary, there are one or two consecutive matches in the syllables in the Pinyin string is located at the last name dictionary matched syllable follow immediately with a syllable name dictionary, and determines a syllable is located next to the name of the dictionary, if located, the match is successful; if not located, the names of two syllables and syllables follow the dictionary next to the match will be compared to determine the two syllables is located next to the name of the dictionary, if located, the match is successful, otherwise the match fails.

[0081] 整句生成模块20进一步用于,在匹配失败时,选择音节对应的普通候选词参与整句的生成。 [0081] sentence generation module 20 is further configured to, when the matching fails, select the candidate word syllable corresponding to general involved in generating the sentence.

[0082] 以上所述,仅为本发明的较佳实施例而已,并非用于限定本发明的保护范围。 [0082] The above are only preferred embodiments of the present invention but are not intended to limit the scope of the present invention.

Claims (12)

1. 一种中文输入过程中的整句生成方法,其特征在于,该方法包括:对用户输入的拼音串进行音节划分,并将划分的各个音节与预设的姓氏字典和名字字典进行比较;当匹配到所述拼音串中存在一个或两个连续的音节位于所述姓氏字典中,且匹配音节后续紧邻的一个或两个连续的音节位于所述名字字典中时,将所匹配的姓氏音节和名字音节对应的人名作为候选词参与整句的生成,并将所述整句生成的结果通过界面显示。 A Chinese input sentence generating process, characterized in that, the method comprising: a user input Pinyin string is divided into syllables, syllables and divides each compared with a preset first name and last name dictionary dictionary; when matched to the presence of one or Pinyin string of syllables in said two consecutive last name dictionary, and match one or two consecutive on syllable immediately subsequent in said dictionary when the name, the surname syllable matched results name and the name of the corresponding syllable words involved in generating a sentence candidate, and the sentence generated by the display interface.
2.根据权利要求1所述中文输入过程中的整句生成方法,其特征在于,所述姓氏字典和名字字典由服务器提供,客户端从服务器下载。 The method of generating a sentence in the Chinese input process as claimed in claim 1, characterized in that the last name dictionary and the dictionary name provided by the server, the client downloaded from the server.
3.根据权利要求2所述中文输入过程中的整句生成方法,其特征在于,所述姓氏字典中包括收集的姓氏,以及每个姓氏的出现概率和对应的拼音;所述名字字典中包括收集的名字,以及每个名字的出现概率和对应的拼音。 3. The method of claim 2 Chinese sentence input generating process as claimed in claim, wherein the occurrence probability surname surnames included in the collection of dictionary, and each of the last name and the corresponding pinyin; includes the name dictionary the collection's name, as well as the probability of occurrence and the corresponding pinyin each name.
4.根据权利要求3所述中文输入过程中的整句生成方法,其特征在于,所述将所匹配的姓氏音节和名字音节对应的人名作为候选词参与整句的生成,具体为:根据所匹配的姓氏音节的出现概率和名字音节的出现概率,计算出对应人名作为候选词的综合出现概率;在生成整句时,按照预设条件选择综合出现概率较大的一个或多个人名作为候选词参与整句的生成。 The sentence generating method of claim 3 Chinese input process as claimed in claim, characterized in that the said syllable matched last name first name and name of the corresponding syllable words involved in generating a sentence candidate, specifically: in accordance with the appear appearance match the name and surname syllable syllable probability of probability, calculate the corresponding name as a candidate word probability synthesis occurs; when generating the sentence, under a predetermined condition to select a larger probability a more comprehensive personal name appears as a candidate involved in generating word of the sentence.
5.根据权利要求1、2、3或4所述中文输入过程中的整句生成方法,其特征在于,所述将划分的各个音节与预设的姓氏字典和名字字典进行比较,进一步包括:将拼音串划分出的各个音节与所述姓氏字典进行比较,在匹配到所述拼音串中存在一个或两个连续的音节位于所述姓氏字典时,将所述匹配音节后续紧邻的一个音节与所述名字字典进行比较,判断所述紧邻的一个音节是否位于名字字典中,如果位于,则匹配成功; 如果不位于,则将所述匹配音节后续紧邻的两个音节与所述名字字典进行比较,判断所述紧邻的两个音节是否位于名字字典中,如果位于,则匹配成功,否则匹配失败。 The Chinese input method sentence generation process of claim 2, 3 or 4, wherein each of said dividing syllables with a preset name and last name dictionary dictionary, further comprising: when the Pinyin string of syllables divided is compared with the last name dictionary, or the presence of a continuous two syllables in said dictionary matching surnames to the Pinyin string, the syllable a syllable match with the immediately subsequent the name of the dictionary are compared, a determination of the syllable is located immediately name dictionary, if located, the matching is successful; if not located, then the match syllable immediately subsequent two syllables and compares the name dictionary judging whether the two syllables is located next to the name of the dictionary, if located, the match is successful, otherwise the match fails.
6.根据权利要求5所述中文输入过程中的整句生成方法,其特征在于,该方法进一步包括:在匹配失败时,选择所述音节对应的普通候选词参与整句的生成。 The method of generating a Chinese sentence input process as claimed in claim, wherein the method further comprises: when the matching fails, the selected candidate word syllable ordinary involved in generating the corresponding sentence of.
7. 一种中文输入过程中的整句生成装置,其特征在于,该装置包括:音节划分比较模块,用于对用户输入的拼音串进行音节划分,并将划分的各个音节与预设的姓氏字典和名字字典进行比较;整句生成模块,用于根据所述音节划分比较模块的比较结果,在匹配到所述拼音串中存在一个或两个连续的音节位于所述姓氏字典中,且匹配音节后续紧邻的一个或两个连续的音节位于所述名字字典中时,将所匹配的姓氏音节和名字音节对应的人名作为候选词参与整句的生成;显示模块,用于将所述整句生成的结果通过界面显示。 A Chinese character input process sentence generating means, wherein, the apparatus comprising: a comparison module syllable division, for Pinyin string entered by the user syllable division, and the division with a preset surname syllables comparing the dictionary and the dictionary name; sentence generation module according to the comparison module syllabication comparison result, there are one or two successive syllables located in the last name dictionary matched to the phonetic string, and matching when a syllable or syllables two successive located immediately subsequent to the name of the dictionary, the matched names of syllables and syllable surname name corresponding candidate word involved in generating a sentence; a display module, for the sentence the results generated by the display screen.
8.根据权利要求7所述中文输入过程中的整句生成装置,其特征在于,所述姓氏字典和名字字典由服务器提供,整句生成装置所在的客户端从服务器下载。 8. Chinese sentence generating means in the input process as claimed in claim 7, wherein the last name dictionary and the dictionary name provided by the server, the client apparatus generates a sentence where the downloaded from the server.
9.根据权利要求8所述中文输入过程中的整句生成装置,其特征在于,所述姓氏字典中包括收集的姓氏,以及每个姓氏的出现概率和对应的拼音;所述名字字典中包括收集的名字,以及每个名字的出现概率和对应的拼音。 9. The Chinese sentence generating means in the input process according to claim 8, wherein the surname appears last name dictionary includes a collection, and each of the last name and the corresponding pinyin probabilities; includes the name dictionary the collection's name, as well as the probability of occurrence and the corresponding pinyin each name.
10.根据权利要求9所述中文输入过程中的整句生成装置,其特征在于,所述整句生成模块进一步包括:概率计算子模块,用于根据所匹配的姓氏音节的出现概率和名字音节的出现概率,计算出对应人名作为候选词的综合出现概率;整句生成子模块,用于在生成整句时,按照预设条件选择综合出现概率较大的一个或多个人名作为候选词参与整句的生成。 Chinese input means 10. The sentence generation process according to claim 9, wherein said sentence generating module further comprises: probability calculating sub-module configured to appear according to the matched name and surname probability syllable syllable the probability of occurrence is calculated as a candidate word name corresponding to an occurrence probability of the integrated; sentence generation sub-module, configured to, when generating sentence, according to a preset condition selection probability of a larger integrated or personal name as a candidate word appears participation generation of sentence.
11.根据权利要求7、8、9或10所述中文输入过程中的整句生成装置,其特征在于,所述音节划分比较模块进一步用于,将拼音串划分出的各个音节与所述姓氏字典进行比较,在匹配到所述拼音串中存在一个或两个连续的音节位于所述姓氏字典时,将所述匹配音节后续紧邻的一个音节与所述名字字典进行比较,判断所述紧邻的一个音节是否位于名字字典中,如果位于,则匹配成功;如果不位于,则将所述匹配音节后续紧邻的两个音节与所述名字字典进行比较,判断所述紧邻的两个音节是否位于名字字典中,如果位于,则匹配成功, 否则匹配失败。 Chinese sentence generating means 11. The input process 10 or according to claim 8, 9, wherein the comparison module is further used for dividing the syllable, the Pinyin string of syllables divided with the surname comparing the dictionary, or the presence of a continuous two syllables in said dictionary matching surnames to the Pinyin string, the immediately subsequent match syllable a syllable is compared with the name of the dictionary, to determine whether the immediately a syllable dictionary name is located, if located, the matching is successful; if not located, then the match syllable immediately subsequent to the two-syllable name dictionary, and determines the two syllables is located immediately name dictionary, if located, the match is successful, otherwise the match fails.
12.根据权利要求11所述中文输入过程中的整句生成装置,其特征在于,所述整句生成模块进一步用于,在匹配失败时,选择所述音节对应的普通候选词参与整句的生成。 12. The input process 11 Chinese sentence generating apparatus as claimed in claim, wherein said sentence generating module is further configured to, when the match fails, the selected candidate word syllable corresponding to general participation of sentence generate.
CN2010101385319A 2010-03-22 2010-03-22 Method and device for generating whole sentence in Chinese character input process CN102200970A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN2010101385319A CN102200970A (en) 2010-03-22 2010-03-22 Method and device for generating whole sentence in Chinese character input process

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN2010101385319A CN102200970A (en) 2010-03-22 2010-03-22 Method and device for generating whole sentence in Chinese character input process

Publications (1)

Publication Number Publication Date
CN102200970A true CN102200970A (en) 2011-09-28

Family

ID=44661656

Family Applications (1)

Application Number Title Priority Date Filing Date
CN2010101385319A CN102200970A (en) 2010-03-22 2010-03-22 Method and device for generating whole sentence in Chinese character input process

Country Status (1)

Country Link
CN (1) CN102200970A (en)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH10124505A (en) * 1996-10-25 1998-05-15 Hitachi Ltd Character input device
CN1313565A (en) * 2000-03-13 2001-09-19 南京亿迪电子商务有限公司 Chinese-character network (Internet) input technique
CN1647072A (en) * 2002-07-24 2005-07-27 卡西欧计算机株式会社 Electronic dictionary terminal, electronic dictionary server, and recording medium
CN101359254A (en) * 2007-08-03 2009-02-04 北京搜狗科技发展有限公司 Character input method and system for enhancing input efficiency of name entry

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH10124505A (en) * 1996-10-25 1998-05-15 Hitachi Ltd Character input device
CN1313565A (en) * 2000-03-13 2001-09-19 南京亿迪电子商务有限公司 Chinese-character network (Internet) input technique
CN1647072A (en) * 2002-07-24 2005-07-27 卡西欧计算机株式会社 Electronic dictionary terminal, electronic dictionary server, and recording medium
CN101359254A (en) * 2007-08-03 2009-02-04 北京搜狗科技发展有限公司 Character input method and system for enhancing input efficiency of name entry

Similar Documents

Publication Publication Date Title
US8457946B2 (en) Recognition architecture for generating Asian characters
Pasha et al. Madamira: A fast, comprehensive tool for morphological analysis and disambiguation of arabic.
CN101203849B (en) Predictive conversion of user input
US20130073286A1 (en) Consolidating Speech Recognition Results
JP3429184B2 (en) Text structural analyzer and abstracts apparatus, and program recording medium
US7983902B2 (en) Domain dictionary creation by detection of new topic words using divergence value comparison
US8831929B2 (en) Multi-mode input method editor
CN104011712B (en) For cross-language query suggestion evaluate query translation
US10210154B2 (en) Input method editor having a secondary language mode
US20100057457A1 (en) Speech recognition system and program therefor
CN101785000B (en) Word probability determination method and system
US7129932B1 (en) Keyboard for interacting on small devices
KR101491581B1 (en) Correction System for spelling error and method thereof
JP5362095B2 (en) Input method editor
JP4701292B2 (en) Computer system for creating a unique phraseology or terminology dictionary from terminology included in the text data, as well as methods and computer program
KR101465769B1 (en) Dictionary word and phrase determination
CN102693309B (en) Candidate phrase querying method and aided translation system for computer aided translation
JP5011751B2 (en) Translation information output processing program, processing methods and apparatus
CN102859515B (en) Split compound words
US8612206B2 (en) Transliterating semitic languages including diacritics
US20140258892A1 (en) Resource locator suggestions from input character sequence
KR20070027726A (en) Systems and methods for spell correction of non-roman characters and words
US8386237B2 (en) Automatic correction of user input based on dictionary
US8655643B2 (en) Method and system for adaptive transliteration
JP2003223437A (en) Method of displaying candidate for correct word, method of checking spelling, computer device, and program

Legal Events

Date Code Title Description
C06 Publication
C10 Entry into substantive examination
C41 Transfer of patent application or patent right or utility model
COR Change of bibliographic data

Free format text: CORRECT: ADDRESS; FROM: 518044 SHENZHEN, GUANGDONG PROVINCE TO: 518057 SHENZHEN, GUANGDONG PROVINCE

ASS Succession or assignment of patent right

Owner name: SHENZHEN SHIJI LIGHT SPEED INFORMATION TECHNOLOGY

Free format text: FORMER OWNER: TENGXUN SCI-TECH (SHENZHEN) CO., LTD.

Effective date: 20131016

C12 Rejection of a patent application after its publication