摘要:
A method for normalizing query words in web search includes populating a dictionary with join and split candidates and corresponding joined and split words from an aggregate of query logs; determining a confidence score for join and split candidates, a highest confidence score for each being characterized in the dictionary as must-join and must-split, respectively; accepting queries with words amenable to being split or joined, or amenable to an addition or deletion of a hyphen or an apostrophe; generating, based on the accepted queries, split candidates obtained from the dictionary, and candidates of join, hyphen, or apostrophe algorithmically; and submitting to a search engine the generated possible candidates characterized as must-join or must-split in the dictionary, to improve search results returned in response to the queries; applying a language dictionary to generated candidates not characterized as must-split or must-join, to rank them, and submitting those highest-ranked to the search engine.
摘要:
A system and method for ranking web searches with quantified semantic features. A query for a web search is received from a user. The query is segmented and tagged into one or more linguistic segments using linguistic analysis. At least some of the linguistic segments are tagged with a linguistic type. A query execution plan is generated comprising the linguistic segments and, for each of the linguistic segments tagged with a linguistic type, at least one tag attribute comprising at least one domain specific feature of the linguistic type. A search is performed for documents matching the query. Each of the documents is scored for each of the linguistic segments of the query execution plan using the tag attributes of the respective linguistic segment. The documents are ranked using a function that uses the scores of the documents. A ranked list of the documents is transmitted back to the user.
摘要:
An aggregate ranking model is generated, which comprises a general ranking model and one or more topical training models. Each topical ranking model is associated with a topic, or topic class, and for use in ranking search result items determined to belong to the topic, or topic class. As one example, the topical ranking model is trained using a set of topical training data, e.g., training data determined to belong to the topic, or topic class, a general ranking model and a residue, or error, determined from a general ranking generated by the general ranking model for the topical training data, with the topical ranking model being trained to minimize the general ranking model's error in the aggregate ranking model.
摘要:
A method is provided for selecting relevant documents returned from a search query. When a search engine finds search terms in documents, the document score is based on the frequency of the occurrence of those terms, the category of the term, and the section of the document in which the term is found. Each (category type, document section) pair is assigned a weight that is used to modify the contribution of term frequency. The weights are determined in an offline process using historical data and human validation. Through this empirical process, the weight assignments are made to correlate high relevance scores with documents that humans would find relevant to a search query.
摘要:
Techniques are described for automatically determining which terms in a search query may be augmented by contextually similar terms such that more relevant results can be displayed to a user. Contextually similar words are determined based on training data, including a web corpus and a query log. Once contextually similar words are determined, they may be inserted into a search query and used to find more relevant results. Consequently, documents that contain helpful information but may not have exact word matches may be found more readily by a search engine.
摘要:
One embodiment selects from a set of query-suggestion pairs a first query and a subset of query-suggestion pairs that each has the first query as its query; computes a Log Likelihood Ratio (LLR) value for each query-suggestion pair from the subset of query-suggestion pairs; ranks the subset of query-suggestion pairs according to their respective LLR values; removes from the subset of query-suggestion pairs all query-suggestion pairs whose LLR values are below a predetermined LLR threshold; computes a Pointwise Mutual Information (PMI) value for each remaining query suggestion pair from the subset of query-suggestion pairs; removes from the subset of query-suggestion pairs all query-suggestion pairs whose PMI values are below a predetermine PMI threshold; and constructs a ranked set of suggestions for the first query, wherein the ranked set of suggestions comprises one or more suggestions of the remaining query-suggestion pairs from the subset of query-suggestion pairs.
摘要:
Techniques for automatically scoring submissions to an online question-and-answer submission system are disclosed. According to one such technique, an initial set of user submissions are scored by human operators and/or automated algorithmic mechanisms. The submissions and their accompanying scores are provided as training data to an automated machine learning mechanism. The machine learning mechanism processes the training data and automatically detects patterns in the provided submissions. The machine learning mechanism automatically correlates these patterns with the scores assigned to the submissions that match those patterns. As a result, the machine learning mechanism is trained. Thereafter, the machine learning mechanism processes unscored submissions. The machine learning mechanism automatically identifies, from among the patterns that the machine learning mechanism has already detected, one or more patterns that these submissions match. The machine learning mechanism automatically scores these submissions based on the matching patterns and the scores that are associated with those patterns.
摘要:
A speech recognition system having an input for receiving an input signal indicative of a spoken utterance that is indicative of at least one speech element. The system further includes a first processing unit operative for processing the input signal to derive from a speech recognition dictionary a speech model associated to a given speech element that constitutes a potential match to the at least one speech element. The system further comprised a second processing unit for generating a modified version of the speech model on the basis of the input signal. The system further provides a third processing unit for processing the input signal on the basis of the modified version of the speech model to generate a recognition result indicative of whether the modified version of the at least one speech model constitutes a match to the input signal. The second processing unit allows the speech model to be modified on the basis of the recognition attempt thereby allowing speech recognition to be effected on the basis of the modified speech model. This permits adaptation of the speech models during the recognition process. The invention further provides an apparatus, method and computer readable medium for implementing the second processing unit.
摘要:
A technique for identifying one or more items from amongst a plurality of items in response to a spoken utterance is used to improve call routing and information retrieval systems which employ automatic speech recognition (ASR). An automatic speech recognizer is used to recognize the utterance, including generating a plurality of hypotheses for the utterance. A query element is then generated for use in identifying one or more items from amongst the plurality of items. The query element includes a set of values representing two or more of the hypotheses, each value corresponding to one of the words in the hypotheses. Each value in the query element is then weighted based on hypothesis confidence, word confidence, or both, as determined by the ASR process. The query element is then applied to the plurality of items to identify one or more items which satisfy the query.
摘要:
Techniques for automatically scoring submissions to an online question-and-answer submission system are disclosed. According to one such technique, an initial set of user submissions are scored by human operators and/or automated algorithmic mechanisms. The submissions and their accompanying scores are provided as training data to an automated machine learning mechanism. The machine learning mechanism processes the training data and automatically detects patterns in the provided submissions. The machine learning mechanism automatically correlates these patterns with the scores assigned to the submissions that match those patterns. As a result, the machine learning mechanism is trained. Thereafter, the machine learning mechanism processes unscored submissions. The machine learning mechanism automatically identifies, from among the patterns that the machine learning mechanism has already detected, one or more patterns that these submissions match. The machine learning mechanism automatically scores these submissions based on the matching patterns and the scores that are associated with those patterns.