Abstract: The present invention relates to a tantalum wire for anode lead of tantalum capacitors, characterized in that the cross section of the tantalum wire is approximate rectangular or regular rectangular. The present invention also relates to a process for manufacturing the tantalum wire, comprising the steps of: providing feedstock tantalum wire; subjecting the feedstock tantalum wire to heat treatment; subjecting the heat treated tantalum wire to surface pretreatment to form an oxide membrane on the surface-pretreated tantalum wire; rolling the surface-pretreated tantalum wire by lubricating with lubricant oil to make the cross section of the rolled tantalum wire being approximate rectangular or regular rectangular; subjecting the tantalum wire to final annealing.

Abstract: Techniques for identifying multilingual cognates and using the multilingual cognates are provided. In one technique, multilingual cognates identified from multiple user profiles are used to train one or more translation models. In another technique, multilingual cognates identified from a single user's profile are used to translate text provided by that user. In another technique, multilingual cognates from a single user are used to align sentences in one language to sentences in another language and the aligned sentences are used to train a language model. In another technique, multilingual cognates identified from multiple user profiles are used to expand search queries. In another technique, multilingual cognates identified from multiple user profiles are used to translate other users' profiles into a target language so that users associated with a source language are viewing the other users' profiles.

Abstract: The present disclosure discloses a digital-to-analog converter (DAC) design which is suitable for providing a high output power high-speed DAC, e.g., in radio frequency applications. The DAC design utilizes a parallel DAC structure, e.g., having 8 parallel DACs and an aggregate current output, to provide a high and programmable current output (in some implementations, up to 512 mA or more). The parallel DAC structure alleviates the design problems which exist in trying to output a high amount of current using a single DAC. The DAC design further utilizes a hybrid structure which integrates the signal chain for a more reliable system. In some embodiments, the hybrid structure uses a CMOS process for the current sources and switches and a GaAs cascode stage for combining the outputs to optimally leverage the advantages of both technologies. The result is a highly efficient DAC (with peak output power programmable up to 29 dBm or more).

Abstract: Provided are a laterally diffused metal oxide semiconductor field-effect transistor and a manufacturing method therefor. The method comprises: providing a wafer on which a first N well (22), a first P well (24) and a channel region shallow trench isolating structure (42) are formed; forming a high-temperature oxidation film on the surface of the wafer by deposition; photoetching and dryly etching the high-temperature oxidation film, and reserving a thin layer as an etching buffer layer; performing wet etching, removing the etching buffer layer in a region which is not covered by a photoresist, and forming a mini oxidation layer (52); performing photoetching and ion injection to form a second N well (32) in the first N well and form a second P well (34) in the first P well; forming a polysilicon gate (62) and a gate oxide layer on the surface of the wafer; and photoetching and injecting N-type ions to form a drain electrode (72) and a source electrode (74).

Abstract: The present invention relates to a method for manufacturing a laterally insulated-gate bipolar transistor, comprising: providing a wafer having an N-type buried layer (10), an STI (40), and a first N well (22)/a first P well (24) which are formed successively from above a substrate; depositing and forming a high-temperature oxide film on the first N well (22) of the wafer; performing thermal drive-in on the wafer and performing photoetching and etching on the high-temperature oxide film to form a mini oxide layer (60); performing photoetching and ion implantation so as to form a second N well (32) inside the first N well (22) and second P wells (34) inside the first N well (22) and the first P well (24); then successively forming a gate oxide layer and a polysilicon gate (72), wherein one end of the gate oxide layer and the polysilicon gate (72) extends onto the second P well (34) inside the first N well (22), and the other end extends onto the mini oxide layer (60) on the second N well (32); and photoetching

Abstract: The present invention relates to a method for manufacturing a laterally insulated-gate bipolar transistor, comprising: providing a wafer having an N-type buried layer (10), an STI (40), and a first N well (22)/a first P well (24) which are formed successively from above a substrate; depositing and forming a high-temperature oxide film on the first N well (22) of the wafer; performing thermal drive-in on the wafer and performing photoetching and etching on the high-temperature oxide film to form a mini oxide layer (60); performing photoetching and ion implantation so as to form a second N well (32) inside the first N well (22) and second P wells (34) inside the first N well (22) and the first P well (24); then successively forming a gate oxide layer and a polysilicon gate (72), wherein one end of the gate oxide layer and the polysilicon gate (72) extends onto the second P well (34) inside the first N well (22), and the other end extends onto the mini oxide layer (60) on the second N well (32); and photoetching

Abstract: The present disclosure relates to a method and a device for controlling information transmission and a smart terminal. The method includes: receiving an instruction for outputting information sent by a near field communication (NFC) chip, an information output function of the NFC chip being in a disabled state and information to be outputted being in an unavailable state; performing authentication of an identity of a user who triggers the information output; and if it is determined that the authentication of the identity is passed, enabling the information output function of the NFC chip and setting the information to be outputted as in an available state, such that the NFC chip can output the information. In the present disclosure, the information transmission through NFC can be more safe and reliable.

Abstract: Described herein are apparatuses, systems and methods for adaptive downlink scheduling and link adaptation. The methods including, at a base station connected to a user equipment (“UE”), determining an initial modulation and coding scheme (“MCS”) for a plurality of subframes to be transmitted to the UE, wherein each MCS relates to a coding rate value for the subframes, determining an MCS pattern for the plurality of subframes based on the initial MCS, wherein an MCS for one of the subframes has a higher coding rate value than the initial MCS, and transmitting the plurality of subframes to the UE according to the MCS pattern.

Abstract: Techniques are provided to allow users of a social network to have multilingual profiles (or profiles in second languages that are different than the users' native, or original, profile languages). In one technique, a translation model is applied to one or more data items (in a first language) in a user's profile to generate translated data items in a second language. The translated data items are displayed to the user (or an online social “friend” of the user) and the user is allowed to select one of the translated data items to include in the user's “second language” profile. The selection may then be used to improve the translation model.

Abstract: Techniques are provided to allow users of a social network to have multilingual profiles (or profiles in second languages that are different than the users' native, or original, profile languages). In one technique, a translation model is applied to one or more data items (in a first language) in a user's profile to generate translated data items in a second language. The translated data items are displayed to the user (or an online social “friend” of the user) and the user is allowed to select one of the translated data items to include in the user's “second language” profile. The selection may then be used to improve the translation model.

Abstract: This application discloses methods and processes for preparation of high purity aluminum hydroxide and high purity aluminum oxide. The method of preparing high purity aluminum hydroxide involves reacting aluminum with water in the presence of one or more catalysts and one or more complexing agents that can react with non-aluminum metal impurities to form soluble complexes for effective removal through rinsing.

Abstract: Techniques are provided for determining one or more recommendations to a user based on job change indications. In one approach, job change indications are tracked to generate associations between job information, such as an association between a user's job title at one time and the user's job title at a later time or an association between a user's job skill and a job skill of a job position to which the user applied. A mapping (such as a translation model) is updated based on the associations. The mapping is used to determine one or more recommendations for a particular user, where the recommendations may be job openings for the particular user, candidate users for the particular user to recruit, suggestions to edit a profile of the particular user, or possible query expansions for a query of the particular user.

Abstract: Techniques for identifying multilingual cognates and using the multilingual cognates are provided. In one technique, multilingual cognates identified from multiple user profiles are used to train one or more translation models. In another technique, multilingual cognates identified from a single user's profile are used to translate text provided by that user. In another technique, multilingual cognates from a single user are used to align sentences in one language to sentences in another language and the aligned sentences are used to train a language model. In another technique, multilingual cognates identified from multiple user profiles are used to expand search queries. In another technique, multilingual cognates identified from multiple user profiles are used to translate other users' profiles into a target language so that users associated with a source language are viewing the other users' profiles.

Abstract: Techniques for identifying multilingual cognates and using the multilingual cognates are provided. In one technique, multilingual cognates identified from multiple user profiles are used to train one or more translation models. In another technique, multilingual cognates identified from a single user's profile are used to translate text provided by that user. In another technique, multilingual cognates from a single user are used to align sentences in one language to sentences in another language and the aligned sentences are used to train a language model. In another technique, multilingual cognates identified from multiple user profiles are used to expand search queries. In another technique, multilingual cognates identified from multiple user profiles are used to translate other users' profiles into a target language so that users associated with a source language are viewing the other users' profiles.

Abstract: A method for manufacturing high-purity alumina material is disclosed, which includes the steps of: reacting aluminum metal with a mixture of organic base and water to form aluminum hydroxide suspension; removing water by filtration to form aluminum hydroxide slurry, and drying/baking the slurry to form aluminum oxide powders. The method is amenable to mass production of high-purity aluminum oxide containing total silica and non-aluminum metal impurities less than 0.005% and having a bulk density higher than 3.0 g/cc. In addition, the invention also provides high-purity aluminum oxide prepared by using the method disclosed and bulk products prepared therefrom.

Abstract: Techniques for training a tokenizer (or word segmenter) are provided. In one technique, a tokenizer tokenizes a token string to identify individual tokens or words. A language model is generated based on the identified tokens or words. A vocabulary about an entity, such as a person or company, is identified. The vocabulary may be online data that refers to the entity, such as a news article or a profile page of a member of a social network. Some of the tokens in the vocabulary may be weighted higher than others. The language model accepts the weighted vocabulary as input and generates pseudo sentences. Alternatively, regular expressions are used to generate the pseudo sentences. The pseudo sentences are used to train the tokenizer.

Abstract: Described are methods that enable rapid automated object classification of measured three-dimensional object scenes. Each method can be performed during a three-dimensional measurement procedure while data are being acquired or after completion of the measurement procedure using the acquired data. In various embodiments, an object scene is illuminated with an optical beam and an image is acquired. In some embodiments, the object scene is illuminated with a structured light pattern and a sequence of images of the object scene illuminated by the pattern at different spatial phases is acquired. Coordinates are determined for points in the one or more images and a translucence value is determined for each of the points. An object class is determined for each point based on the translucence value for the point. Optionally, additional information, such as grayscale or color image data for each point, is used to supplement the object class determination.

Abstract: Techniques for training a tokenizer (or word segmenter) are provided. In one technique, a tokenizer tokenizes a token string to identify individual tokens or words. A language model is generated based on the identified tokens or words. A vocabulary about an entity, such as a person or company, is identified. The vocabulary may be online data that refers to the entity, such as a news article or a profile page of a member of a social network. Some of the tokens in the vocabulary may be weighted higher than others. The language model accepts the weighted vocabulary as input and generates pseudo sentences. Alternatively, regular expressions are used to generate the pseudo sentences. The pseudo sentences are used to train the tokenizer.